Previous MageLair News
9/4/01
Rewrote the Delayed Startup program. A systray icon is now displayed while the program is running and you have 3 menu options (About, Edit, and Exit). Note, any task that was already executed will be re-executed if you select edit (regardless of whether or not you change the file) as the configuration is re-read and all tasks will be flagged as not run (just as they would be if you just ran the program). I may, at a later time, change the code so that any task that should have already been started will be auto flagged as executed (based on the current system uptime when the configuration file is being read back into the array).
8/30/01
Did some testing on the CPU and motherboard that was pulled from the MageLair server. I suspect a bad motherboard, but until I get some extra PC-100 or PC-133 DIMMs, I won't know for sure. Since I came to a stopping point on the hardware problems, I decided to do some tweaking in my Caller*ID program. The only thing really new is that it monitors the window position and keeps it within view. It doesn't perform this check when first loading, but I don't think that will be a problem as it should be saving valid coordinates anyway. I'm sure you can find your way to download the current Caller*ID package (source and executable files still included). If not, look on the left side of your browser window. :)
8/29/01
Due to an unforeseen problem (2 consecutive power supply failures, coupled with a possible motherboard death), the main MageLair server has just been downgraded to an Intel Celeron 300. Until a faster CPU can be re-installed into this server, the Quake 3 server will be offline. I hope to have more details in the next day or two.
8/28/01
For those that haven't noticed (and who would that be?), I just finished changing the site a bit to include the use of frames. I know, it's old web technology and nobody's using that anymore (seems everyone is moving to more advanced methods that aren't supported by all browsers, not that frames are supported by everything but I don't see anyone coming here via lynx). I thought about using a CSS (aka cascading style sheets) configuration, but decided I wanted to keep the site Netscape friendly (for now, anyway). As I take the time, I will be adding more features to the site. May toss in a few graphics to replace some text hyperlinks and such, but I don't know.
8/24/01
I have a new program available. I wrote it due to an odd behavior with the SETI@Home command-line client on my system. If I have the client start in my StartUp group, the system runs extremely sluggish. That's bad when you consider that it makes an Athlon 1.5GHz system act like it was a 486. But, that wasn't the only reason I wrote this program. A friend of mine asked for a program to auto-start mIRC a little after Windows finished booting. I don't know her exact reason, but when she said she wanted something like that, well, I needed to do it anyway. So, Delayed Startup was written to fill this gap. In that archive you'll find a precompiled executable, a sample configuration file which must be stored in the Windows directory (i.e. C:\Windows), and the C source. Once all defined "tasks" have been launched, my program will exit. However, if you entered something wrong, you'll have to use a task manager to kill the program as I have no user interface coded into this thing. No icon loads to say it's there and no window shows up anywhere to say it's running. Feel free to add a user interface if you like. Make it a systray app with built-in configuration editing if you feel the need. All I ask is that you send me a copy of whatever you do with it.
8/15/01
Just finished setting up the Quake III: Arena server. I'm still sifting through the maps, removing duplicates and maps too small. If you find any that I have missed, please email me and I'll get it taken care of. If you want any special maps/skins/models added, email me about it and we'll talk.
8/13/01
More Caller*ID code updates. Added some more code to try and catch erroneous data packets. Also moved the code that saves the call history into the registry out of the SaveOptions() routine into it's own SaveHistory() routine. With that, I have the AddToHistory() routine call SaveHistory(), so the calls that occurred after the last proper exit (due to crash or power outage) won't be lost as they have been. We'll see if I still have the problem with noise causing a memory access violation or if that's now fixed. I wouldn't recommend anyone hold their breath.
8/11/01
Upgraded the server today. It was running an AMD K6-2 @ 300MHz and was working fine. However, with wanting to run a Quake 3 server on it, I felt that upgrading was needed. It's now running an AMD Athlon @ 700MHz I also took out the SCSI controller and the SCSI CD-ROM drive as they were not being used. I should have the Quake 3 server running any day now.
7/31/01
I just uploaded a possibly fixed Caller*ID source/executable package. There's been 2 bugs that have remained all this time. One of them involved the TCP/IP client/server handshake. If a password was specified, the server would simply lock up. I have looked at the code involved in that check and added something that I hope will fix the problem. I have not yet tested the code change, so it may yet remain broken. The other bug is more obscure and less likely to be triggered. There's a bug (hopefully fixed now) in the data stream processing code that would cause a memory access violation and terminate the program. The only cause of this would be garbage data that happened to look like a Caller*ID message packet, but contained invalid length bytes. I had added code to catch this, or so I thought, but looking for values larger than the allocated buffer space. However, I didn't think to check for the length to be less than 1, so a 0 value could slip in there and really cause trouble. Now, all fields that are variable length and checked for a length of at least 1, but less than the buffer (which is based on the largest data length as specified by a Caller*ID document that I read). And all the fields that are a fixed length (like the date/time) are checked for that specific length. If either of these checks fail, the entire packet is dropped as garbage.
7/25/01
I received an ATI Xpert 128.(Rage 128 w/16MB and PCI) about a week ago. I decided that I'd rather not downgrade the video in my Athlon (GeForce 2 GTS w/32MB and AGP), so I see about updating it's BIOS for use in my PowerMac. After searching the web, I found little hope of such an existing fix. Well, not wanting to give up, I sent ATI's tech support a message asking for help. Today, finally, I got a response. Here's what they had to say:
"Regarding Xpert 128 PCI for a PC be used in a MAC:
The BIOS on the Xpert 128 is cannot be flashed with an software update. The BIOS on PC and MAC products are specific to their respective environments. The Xpert 128 PC card you have cannot be used in your Power Macintosh 7500.
Regards,
Edwin Abuyuan
Technical Support Canada
ATI Technologies, Inc."
Now, the funny thing is, the ROM on the Xpert 128 is an Atmel AT49F512. That's a 64k, 5v only, flash memory chip. You can view it's details here. Now, what's all this about it "cannot be flashed with an software update" all about? Looks to me like it sure can. That is, unless they pre-flash the chip, solder it to the board, and have no signal going to the write enable line. Seems stupid to me. I mean, why use a flash memory chip if you don't have the circuit support it's capabilities? What if they ever need to release a fix for something in the BIOS? Somehow, I don't buy the "cannot be flashed" story. The funniest part was that I asked about an exchange possibility (provided it wasn't going to cost more than the card was worth) in the case of no flash "fix" being currently available. They ignored that request. It seems to boil down to ATI not caring anything about the consumer once they have your money. Nice strategy, huh.
7/2/01
Updated the Quake information as listed above. I hadn't played Quake2 or Unreal Tournament in a long time and decided to go ahead and setup a semi-permanent Quake3 server on my router. With on a 256k upstream, I can't expect it to handle heavy loads (several users) without lag. Once I get sufficient CPU power in the magelair.com host, the permanent server will be run there.
6/21/01
I have a correction. The W6-LI does not require ECC or EDO. I would prefer EDO over FPM, but I won't refuse the FPM modules. So, what I'm looking for is 3.3v buffered 168-pin DIMMs. Thank you.
6/20/01
It's been a while since I updated. Anyway, I'm preparing to upgrade the this server. It's one of those older hardware but faster result situations. A friend just dropped off an old Micronics W6-LI board. Now, considering I'm on PC100 memory (SDRAM), why would I want to drop an old FX chipset motherboard into this server? It's really quite simple. The W6-LI is a dual Pentium Pro board. A single AMD K6-2 @ 300MHz, or even a K6-2 @ 500MHz as the plan was to upgrade to that, still wouldn't beat the performance of a pair of Pentium Pro Overdrive processors. Especially if those CPUs were each running at 333MHz. I haven't decided, yet, but I'm thinking of still running just one SETI client on the server. Why, you ask? Because I will probably install and run a full-time Quake III: Arena server on the other/extra CPU. At this time, I am preparing to purchase the CPUs for the board. However, I feel that the prices for RAM that this board takes are a bit high. So, I'm asking for donations to cover this problem. I'm not asking for money. What I'm looking for is 3.3v buffered ECC EDO DIMM memory donations. I would prefer 64MB and 128MB modules only. The server is currently running on 192MB (1x 64MB and 1x 128MB) and I would like to, at the very least, keep that much RAM in the system. Anyone interested in donating can email me and we can make arrangements from there. I will be putting together a page that details where various hardware came from, so you would get credit for anything you donate. Again, if the RAM I needed for this board wasn't so ridiculously high priced, I would just buy it. In fact, if anyone can show me a place to buy the 3.3v buffered ECC EDO RAM for roughly the same price as PC133 memory, I would greatly appreciate it.
4/26/01
I switched myself to Windows 2000 Professional this week. It's proven to be just as compatible and more robust/stable than Windows ME. I've even been able to load WindowBlinds and WindowFX without worrying about resources vanishing. Sadly, I discovered that Windows 2000 doesn't allow my Caller*ID and LCD programs to add/remove themselves from the Run key in the registry. The simple code to perform this task in Windows 9x/ME seems to be insufficient. Once I find the correct code block to perform this task, I'll get those programs updated.
On a different note, a very close friend tried to order a motherboard/CPU combo from www.bzboyz.com. The combo was for an Asus A7M266 and the AMD Thunderbird 1.3GHz. Well, the incompetents at www.bzboyz.com called to say they could fill the order because the A7M266 was a discontinued product and that they were going to replace it with the A7A266. Knowing that the A7M266 is not discontinued, I had this friend call them back. They still claim the A7M266 is a problematic board and has been replaced with the A7A266. It is my belief that the idiots at www.bzboyz.com really don't have a clue, are simply out of stock, and chose a very bad lie to cover themselves.
4/6/01
I recently completed a small project that I'd been thinking about for a few months. You may have noticed the program I wrote a while back that used an LCD display hooked via the parallel port. Well, that program was nice, but it lacked many features that I had added since I posted it here and it wouldn't work in a WinNT/Win2k configuration (no direct access to the parallel port). Now those issues are behind me. I have here the archive containing the schematic diagram, PCB layout, source and object for the Atmel microcontroller, and the C source for the Windows application. A simple, 3-wire serial interface is all that is used (and I don't even use one of them, myself).
3/5/01
Added the magelair.com server to the list of systems running SETI@Home. Also, I've put up a rather basic web site for the group, located here.
3/3/01
Formed a SETI@Home group today. I have 4 personal systems running the software (2 Linux systems, my Windows ME system, and my PowerMac). Don't know what SETI@Home is? Check the link at the bottom of the page.
1/20/01
Had a fun time getting the server hacked early yesterday morning. Had things fixed back up within 4 hours of the system being taken over, but I won't say it's over. It's been said that there are always ways into a system. The more users you have on a system, the easier it is to get in. The more services your server runs, the more vulnerable you are to attack. The only truly secure system is a computer with no mouse, keyboard, or power cord. But, what use would that be?
On a happier note, I've been redesigning the SmartIDE interface. I've added an 8k SRAM that gets banked in at $D600-$D7FF and made the battery backed SRAM only enabled in place of the original OS ROM. I'm still looking into making the changes for the SmartIDE conversion to PBI specification.
1/13/01
I just put my v3 RAM disk code, for the Atari, on the site. I don't recommend this code for use. Not that I had trouble with it, but more because of the method used for determining the available RAM banks. A 2k SRAM and a PIA @ $D700-$D703 are required by the code as it stands.
1/10/01
Have you ever wondered what kind of mess I have for a computer area? Well, here is a pic of just a portion of the room. This doesn't include my workbench/soldering station. It doesn't show the whole desk with my PowerMac. Nor does it show my storage closets and table. There's also a Blasteroids arcade machine and fairly decent entertainment center (stereo, 7-disc CD, satellite, VCR, DVD, and 25" TV). But, I don't feel like showing the entire room in it's current state. I have a little paperwork problem (the problem is that it's all over the place). Anyway, you'll notice letters all over the picture. I placed letters on or near key elements of my main computer area. You can either email me about one or more of them, or wait for me to post another page with descriptions of each of them.
1/1/01
I've been working on a major expansion to the SmartIDE project. I'll post a work-in-progress schematic sometime in the nest few days. It's mostly complete, but I still have some minor details to work out.
12/19/00
Evening: My 2nd 74ALS573 was not bad. I had pulled these chips from a bad Pentium motherboard a while back, and the method I used left excess solder on the pins. I had a short between at least 2 pins on my 2nd chip. After cleaning it up and adding some code to my IDE source (modified from Bob Woolley's original IDE38.ASM), I had verified full read and write functionality of my modification. Please visit my Atari page to obtain the source and a .BMP image of the schematic changes (I, again, apologize for the poor quality of the schematic). Currently, since I doubt any DOS for the Atari really handles 512 byte sectors, the source adds "r", "w", and "p" for 512 byte sector operations. They read/write the full 512 byte sector, no check is done for desired block length. However, they should otherwise work fine.
Pre-dawn: I have constructed my 16-bit IDE circuit modification. The bad news, I had a bad 74ALS573 and was getting a solid green screen upon turning the system on. After double checking the circuit, I concluded it had to be a bad chip in the read circuit. The good news is that when I pulled the bad chip out and replaced it with the other chip, I do get full 16-bit data reads. For more information, go here. I have also added a few other items to my Atari page. Check it out and feel free to let me know if you like what you see/find.
12/18/00
Naughty me. I didn't do a very good test when I changed the code used in the ATASCII dump. It's now fixed. Also, while I was in the code, I cleaned up a few things. It used to set the left margin to 0 and background color to black. I don't do that anymore. I have a patched OS that does that for me, so I took out the waste of space from the program. Besides, some of you may want to use other color settings. Also, I changed the search command from "S" to "?" and added the single-step and trace functions (using commands "S" and "T"). The code size, obviously, did not go up much, but some good functionality was gained. Also, an error that was causing the saved registers to be written incorrectly has been fixed. Code size is now 2163 bytes. I only wish PC programs with this much power could be written with as little code.
On another note, I have been looking into the possibility of using the full 16-bit I/O for the SmartIDE interface. I do believe that I have found a way to do it with the addition of 2 74573 (octal, tri-state, transparent latch) chips. I intend to set up the circuit in such a way that you read in a low-byte, high-byte order, but write in high-byte, low-byte order. Since the circuit is already there to function on the low byte, the latches would have to be slaved to the existing circuit. If this works, I might be able to access full sectors. Next I'll have to worry about a DOS that supports a 512 byte sector. In theory, SpartaDOS is supposed to (basing my theory on documentation of it's filesystem). However, until I have the ability to read/write the full 512 byte block, I won't know for sure.
12/16/00
Did some more code tweaking with my debugger/mini-assembler. I've removed command's R, W, I, and *. I've also removed the CIO call for the ATASCII display during the hex dump. The ATASCII display is still there, I just remembered there was a flag to make the standard output act the same way I had the CIO call acting. Overall code size is down from 2281 to 2021 (the files, of course, had the standard 6 byte header, but I'm not counting that at this time). That's 260 bytes of unnecessary code removed. My self-test replacement, which doesn't include the mini-assembler, has had the ATASCII dump removed, as well as other tweaks. It's now a puny 1532 bytes. No need to reduce it's size further as that leaves the 0xD6xx and 0xD7xx pages on the SmartIDE wide open for use. With the empty space found in the XL/XE ROM @ 0xCB65-0xCBFF, and dropping the international character set, I have a grand total of 1691 bytes of code/data space available. I have my LCD and SmartIDE code starting at 0xCB65, with the DRVCFG and HDTBL space residing at 0xCFB0-0xCFF0. That leaves me 239 bytes free starting at 0xCEC1. And I still have 0xD600-0xD7FF available (another 512 bytes).
12/13/00
I've been doing a lot of Atari 8-bit coding lately. I've been working on modifying the SmartIDE code from Bob Woolley, changing the location where it resides and adding support for the LCD display modification I made to his circuit. The SmartIDE information can be found here, a web page maintained by Clarence Dyson.
11/13/00
Yet another LCD display program update. No merger with the Caller*ID code, as of yet. Anyway, comment lines can now be added to the config file. I am considering developing the code and schematics for putting together a USB device for both Caller*ID and LCD display. The development kits for USB are not free, and programming devices for USB micro-controllers are rather expensive. So, unless I suddenly get a load of extra cash that I don't have a major use for, or some of you decide to be generous and make some donation for the cause, that project will likely sit on a back-burner for a long time. I would, of course, really like to have these devices running via USB, so I will continue to look for cheaper ways of developing and programming the USB controller chips.
11/7/00
Uploaded another LCD display program update. It wasn't re-reading the configuration from the registry quite right. Also, the Caller*ID code is up to v1.80p2. I may be merging the 2 programs soon, but small/narrow LCD displays will likely be unsuitable for such a merger.
10/30/00
Not much going on, obviously. I updated the Unreal Tournament server and now it doesn't want to run as I had it before. As little as I was using it, I will likely remove it and set up a Q3A server in it's place. Also, updated Caller*ID server code (v1.80 now available for download) with new serial I/O handling. This update has improved functionality on my own system and may be more suitable for multi-line environments. Also, uploaded newer LCD display program (with doc file to describe format for info file).
10/3/00
NOTE: I did not mean to say that the Civil War was crap. The pages that were on this site were not of my design and could have been done much better. The spelling and grammatical errors on those pages were reason enough to remove them.
10/2/00
Civil War crap has been pulled off the site.
7/31/00
Being currently unemployed, I took the time to complete a couple projects that I had started a while back. I finished building a portable ISO-7816 (smartcard) reader/programmer (it plugs into a PS/2 port for power and a 9-pin serial port). I also assembled the SmartIDE project that Bob Woolley designed for the old Atari 8-bit system. He wrote the article with a 1200XL in mind, as that's what he built it in, I assume, so I had to locate the various connection on the main board myself. This was not too difficult, but was fun adding the appropriate lines with all the other extras that I already have in my 130XE. Either way, I have a real conversation starter here. When someone asks me why I did it, I can say "because I can." They'll not understand that, and so I'll have to explain the whole story. :)
7/8/00
After setting the cache to 1/3rd the core speed, I still had rare problems. Bumping the core voltage to 1.95v seems to have cured the remaining problems. Diablo II was the lucky game that showed the instability. I no longer have any troubles.
6/25/00
With my Arctic Circle installed, I decided to give 900MHz a try. With a core voltage of 1.9v and a cache divisor of 2/5, I am currently running at 900MHz. Considering that I have a .25 700MHz core and 3.3ns cache, I feel quite proud of my success. Yes, you read that right. My cache is 3.3ns (303MHz) running at 360MHz (40% of 900MHz). I think I need to get a digital camera and take some pics of my system, just to show what I've done to achieve this feat.
Update: I have reduced the cache speed to 1/3rd that of the core speed. Having the system reboot in the middle of Q3A and refuse to run WinZip indicated certain instability. Now we'll see if we can maintain stability.
6/24/00
Arctic Circle installed and working fairly well. It has reduced my CPU temperature by a minimum of 5 degrees. I am satisfied with the knowledge that I did not waste money. :)
6/22/00
Ever in the quest for faster speed, I have again attempted to reach 850MHz on my Athlon. With a core voltage of 1.85v, I seem to have achieved that goal. Dare I push towards 900MHz on this .25 core? Not until my Arctic Circle arrives. Of course, having just completed several rounds of Q3A and only measuring 45C/113F on the CPU heat, I would say that heat is not going to be much of an issue for me. I do, however, hope to see the Arctic Circle pull that down at least a little bit. Alas, I shall have to wait yet a few more days for the good ole US Postal Service to deliver my cooling upgrade. Previous temperature readings posted here were taken from a thermistor that was secured to the heatsink where it was farthest from the CPU core. Now, this same thermistor is secured on the back side of the heat transfer plate, just above the core. Much more accurate readings. Funny, however, how the heat doesn't seem to go up much during games anymore. This may also be attributed to an extra fan mounted on the side panel of my case, pulling air into the case and blowing it right on to the CPU. I only hope I don't discover that I am already cooling as much as the Arctic Circle. What a waste of money if that be the case.
5/24/00
It's sad when you consider a 32MB AGP video card that is less than 6 months old to be obsolete. I just replaced my Leadtek WinFast GeForce DDR with an ELSA Gladiac (GeForce2 GTS). Granted, I did gain 1200 points in 3DMark 2000, but was it really necessary? No, but when you are a geek like me, you tend to do things because you can, not because you have to. :)
5/22/00
Site updates in progress. I'm starting to debate on using frames (and more java, whether it be applet or script). I need to get some good sounds to use for various things (such as mouse-over effects and the like).
5/20/00
After tearing my Athlon apart (to see the core and cache speeds) I managed to
put it back together more efficient than it was originally built. I have
reduced it's heat output by ~7F. I attribute this to a) better contact
from the cache RAM to the heat transfer plate and b) possibly better heat
transfer compound. About the cache RAM, I have seen it stated (and I saw
it for myself) that the heat transfer plate doesn't make contact with the cache
chips. Using a pair of needle-nose pliers and one of the very thin
backplanes (the kind that usually snap into place) I created a couple shims to
fill in the empty space. It was a perfect fit. Seems that a heat
transfer plate works better when it's getting the heat spread around a bit more.
Site updates are bound to pop up here and there. If you don't like them,
tough. :)
5/13/00
Being almost 1 week after successfully setting my Athlon to 800MHz, I decided
to give 850 another go. With a core voltage setting of 1.85v, I am
currently running at 850MHz. I will likely be changing the resistors on
the Athlon for a cache speed of 1/3 and then trying 900MHz (as soon as I get a
water cooled solution built). At this time, I run ~105F in Windows mode
with 115-120F during heavy gameplay. Although these temps are considered
low, I would prefer lower (by about 20 degrees, if at all possible).
Update: Unstable performance forced me back to 800MHz.
5/7/00
I like living life on the dangerous side. "Why does he say that", you ask? I recently downloaded a little program called FixL2Cache from WDSoft. It modifies the BIOS code for the Asus K7V (Slot A, VIA KX-133 motherboard) to set the Athlon's L2 cache speed. My Athlon 600, with it's default 1/2 core speed setting, would not run stable beyond 750MHz. Using FixL2Cache, I changed to a 2/5 setting and am now running 800MHz. I have run a couple tests, but I have not yet had any problems. I will, of course, post a gripe entry if it decides to mess up.
5/6/00
Being the risk taker that I am (running Microsoft products), I installed IE 5.5 beta. So far, things are working fine. In fact, I now get java applets again. I haven't really had much time with IE 5.5 to really tell how stable it is, so expect some sort of gripe entry before too much longer.
5/2/00
Only with a Microsoft product. Anyone else, you reinstall and things work. Not so with Internet Explorer. I recently discovered that my installation of Internet Explorer was damaged (not running java applets). So, I reinstall. No help. Ok, I had the Microsoft System Information program "repair" the installation. That made things worse. Now, I've told IE5 to reinstall itself again. Still no java, but at least I can browse the web again. It's things like this that make me decide to download and install programs like Netscape Communicator. Now there's a program that I thought I'd never use on my Windows 98 system. But, what else can one do?
4/26/00
Finally got my Athlon upgrade installed and working. I'm ashamed to admit that it took a week for me to figure out that AGP Fast Writes was my whole problem. Funny, though, because I thought that the GeForce DDR would appreciate it, and since the KX-133 chipset is supposed to support, I turned it on. Neither Windows 98SE nor Windows 2000 Professional liked the AGP Fast Writes setting, but RedHat Linux 6.2 installed and ran just fine. XFree86 v3.3.6 even liked the system. And people wonder why I mostly run Linux on my laptop.
4/9/00
Just doubled the background music selection. I now have the MIDI files from Descent, Descent II, and DOOM (episodes 1, 2, and 3) being randomly selected and played.
3/22/00
Ok. So, I had coded the port display code in such a way that on a successful port close, it closes the window and clears the auto-open flag. Ooops. Ok, so saving options will now set a flag to bypass that part of the close sequence, unless the port is no longer set to be active. Let's see what else I can finally get fixed.
2/27/00
Fixed a small bug in the Caller*ID code. Seems I had a minor problem while saving the options where it wasn't doing it right. There is still a minor glitch during the saving of options where it will close the currently opened port displays. Not sure why it does it, but how often does one have to change their options anyway.
2/26/00
Caller*ID code seems pretty stable, so the new stuff is now available on this site.
1/23/00
Minor updates to Caller*ID code and major updates to LCD code still in progress. Caller*ID program still has a problem with the COM I/O threads (they still seem to die after a period of inactivity). LCD program is getting some definable functions (still working things out, but have been busy with other things). To those keeping up with my projects, I thank you for your interests.
1/8/00
Ok. So, it's 8 days into the new year. Now, many will say that this is not the new millennium, and that the new millennium starts on 2001. However, I have one question for those people. Were you born the age of 1 year old? Sorry, but if we are 0 years old during the first year of our lives, I don't see how we can say that the year 2000 is not the beginning of the 21st century. But, hey, I'm a programmer and so I am 0 based in many areas. :) In any case, enjoy your year 2000, whatever century/millennium you think it belongs to.
12/31/99
Here we are, 2 and a half hours away from the ultimate Y2k test. :) Like we wouldn't have heard something from Europe by this time. Unless, maybe it's all true. They can't tell the world because everything blew up, crashed, or simply died. Right, and if I really believed that, I wouldn't be still on my computer, I'd be running for the nearest cave. Recently stumbled onto an old copy of the MageLink source code. This code is from around July of 1995, so it is missing a lot of protocol fixes and feature enhancements. However, it will make a good start for a TCP/IP version. Now I just need to sit down and decide to do the code.
12/23/99
Well, today was a fun day. On the fun side, I got my GeForce card today (gotta love it). The down side, however, my server was the unfortunate victim of being exploited. At least they were nice and left the system intact, aside from trashing the log directory (an obvious way of hiding their trail). Suspected security hole has been plugged, so I guess I wait and see if they leave me any more useful notes. Thanks, whoever you are. I did need to be shown that hole.
11/15/99
Oooooops.....Caller*ID code was only updating the default window coordinates if/when you opened/closed a port display. Fix it now. heh
11/14/99
More Caller*ID project updates. TCP/IP communication between server and client should be fully functional.
11/11/99
Updated Caller*ID project a bit. Client code has been tested and is known to work using semaphore method. TCP/IP communication will be tested in the next couple days. Perhaps I'm all but done. I only know of a problem where I lose contact with the serial port (still trying to figure that out as it doesn't happen at work). No progress has been made with the NIC in my laptop. But I have not given up all hope.
11/6/99
Installed FrontPage 2000 on the laptop. This update is being made with said laptop. The last week I've spent more time playing with and configuring this laptop. I have Win98SE and RedHat Linux 6.1 both on here. The Linux part was easier to get fully operational (minus the NIC) than the Win98SE part. Seems I need a little more work on the pcnet_cs module or the 8390 module and then Linux will be complete. Win98SE, however, had some damaged files from word go, it would seem. But, reinstalling key elements seems to have it fully operational now.
10/26/99
Installed the new motherboard and RAM in the server last night. Just got FrontPage server extensions fixed, too. This is my first update to the site using FrontPage 2000. Back to real work now (shouldn't be editing my website from work, but I had some time to kill).
9/11/99
Resorting to manual web updates. Installed Apache 1.3.9 and FrontPage Server Extensions v4, which worked great the first few times. Now, like all Microsoft programs, it decided to stop functioning. Luckily, I know enough HTML to get the job done.
9/3/99
Happy Birthday to me! :) Anyway, now that I've said that, I've installed a new hit counter. I'm using the LiveCounter Classic java/cgi setup. So, don't be surprised if the number changes every 10 seconds. Not that it will, but in case I suddenly become popular and start receiving lots of visitors.
9/1/99
PlanetQuake should now be getting told about my servers. The problem I had before appears to have been local to my Windows system.
8/28/99
As you can see from the above information, the Q2 servers have been set back up. Due to my LAN configuration, however, I can either have them setup where I can advertise them via PlanetQuake's master browser, or I can leave them as they are and play on the local LAN. I choose to play on the local LAN. If I can find out how to set them up for both, I will. If anyone knows how to have it work properly in a server with 2 NICs, please email me.
8/27/99
Cleaned up a few things. Killed all the old "news". Changed the link for the Q2 clan information to point to the page(s) by my brother. Updated/corrected some hardware information on my about me page. I still need to reconfigure my Q2 server. It was lost when I had a bad DIMM ruin my ext2 filesystem. I salvaged all the important stuff, so I just need to get it going again.
8/26/99
Just moved the site here. Will have to do some updates and other changes eventually.