Older blog entries for Svartalf (starting at number 24)

Just now looking at the work needed for the DMA patch for Utah-GLX. It's actually rather simple (a one-liner in the right place...)- I should have something worked up shortly. Once I've done that, I think that sometime next week I'll go over to the storage space, dig out my Trident documentation and maybe start the work on a possible DRI type driver for my laptop. I've got very little better to do right at the moment- and while it won't put money in my pocket, it might get me noticed again and it'll definitely help get my mind off of my woes.

I was skimming the phazzar discussion list (Phazzar is another attempt to make a next generation Photon system...) and read an article from a fairly reliable source that the original Photon system used a 300 baud IR system initially and then a 1200 baud system. This translates into a single byte code (probably w/parity to do a little error checking) for the individual pack. I was going for overkill (part of the reason for the delay on Tux-Tag was that I was looking for high-speed IR operation, the higher, the better.) when I really ought to have been gunning for dead simple. Given that this is the case, I could make a case for the most bog-simple emitter/detector design and expect it to work like a gem. I also found out that the central computer would poll periodically for scoring info from the pack- apparently, the packs would hold onto who was "hit" by the gun and upload the bad news to the central server, which would then tell the packs in question that they'd been hit and to act accordingly. I think I can do better on both counts (I ought to be able to do better- it's amazing how primitive these systems actually are...) and do it pretty economically. I hope to be drawing up plans for the gun, pack emitters, etc. shortly and start into coding- again, like the almost dead website says, no promises- yet.

Nothing is really happening on the Ballistics front right now- worrying more about some of my Open Source projects. I looked at the sound code and might have some traction on that front shortly. If I do, I'll have something to demo at the presentation I plan on giving to NTLUG in the near future.

Again, there's been some delays, as you can see. I don't have much to blame the delays for my projects on other than my state of mind brought about by me still not finding a job elsewhere and my current so-called employer's still not having any money to pay me with. Yes, you read that quite right. No money so far as I can tell. The current story now from the VC is May 9th. Supposedly it's a done deal at this point and it's just a matter of logistics- somehow, I don't believe them anymore unless I see it. If we don't see some money real quick all my jockying to keep everything is going to be for naught and I'm going to end up in a worse position than I would have been in had I not tried to fight to keep the house, etc.

Do you know of anyone with an open software engineering position? Why don't you send them my resume for review. I'm really good with system level apps, network apps, device drivers, and general consumer apps- on both Windows and Linux. I promise I won't get in anybodys way (much...).

Wow...I'm putting an actual regular diary entry in for a change. A lot of things to talk about- but I'll try to keep it brief...

VC money's s'posed to be this week- signs are such that if it doesn't happen this week, it'll be next for the bridge and perhaps even better yet, the first installment of actual funding. I'm still going to believe it only when I see it- but like I said last week, I think the worst part may be finally over- one way or another.

I haven't heard from the company I interviewed with- but it's very probably too soon for that anyway. I suspect if I'm going to hear anything it'll be by around Friday.

If you're in Texas, there's THREE very important pieces of legislature being discussed in Austin. SB 1579, HB 2121, and SB 1116. SB 1579's a bill we in the state of Texas want, the other two are something we don't want- SB 1579's the bill requiring Open Source to be considered for use in IT projects, the other two are the Texas versions of the so-called S-DMCA bills that RIAA is trying to foist off onto us. Slashdot, the EFF, and others are covering this in much, much more detail, but suffice it to say, it's come time for all the tech crowd to get involved with this stuff if they're not already so.

Got more work done on Ballistics. I'm thinking I might tackle the sound support next because it'll be more impressive for the demo for NTLUG that Michael Simms gave me the green-light on.

Didn't get the patch done for Utah-GLX- ended up doing other work on other things. I plan on tinkering w/it after I verify the tuning work I just did on Ballistics.

Well, boys and girls, it's another fun-filled blog/diary entry for moi- hey, I didn't take more than a couple of months from the last one this time (a definite improvement...)!

Well, guess what, I didn't lose the house- yet. Seems there's some legal angles that I hadn't exploited, so we're still here- for now. But, keeping it (and a few other things...) is contingent on my getting a new job or the current one getting the VC money. Why, you might ask, am I even putting up with these people? Because up until October of last year, I at least had health insurance (More on that one in a little bit...) and I was getting more money than unemployment would have paid for longer- it was just inconsistent. That, combined with this sudden and largely complete discontinuance for any hiring in the industry because of the Internet Bubble bursting, has kept me where I am. Having said all of this, there may be light at the end of the tunnel and it's not the oncoming train. It looks like in 5 or less business days, the VC's going to actually get us our bridge funds as an advance on the first installment of our multiple millions they're funding us. (Please note: I did indicate that it was a crapshoot in the last entry as to when- that was nearly TWO months ago. Waiting for VC's to get their collective act together sucks bigtime...) Couple that with the very real prospects of a new job elsewhere (I recently interviewed with a major corporation that was interested in doing embedded Linux related development- they seemed to be fairly interested in my working for them. I'm not counting my chickens, but it looks good from where I'm standing right at the moment...) and things are possibly looking up for a change.

If you don't have health insurance, you may want to go way, way out of your way to get it however you can possibly do so. At the end of last month, my wife was admitted on an emergency admit to the local hospital for extreme abdominal pains. 3 days in the hospital with nothing more than an IV so that her GI tract would settle down. Her last unemployment check saw to it that we were ineligible for any financial assistance from the county or state. The total bill so far: $10k and rising. Thankfully, we've worked out a solution to the bill and everything will end up okay in the end. Keep the bill in mind, though- you will not be able to work things out everytime like we did with this little visit and you'll be stuck trying to figure out how in the hell to pay it.

The little bit I've done on the Ballistics port has been checked in (and revised several times...). Progress is moving along but we're delayed by things like my trying to keep a house over my head, etc.

On the front of my open sourced projects, I will be submitting a patch to the current Utah-GLX developers to fix the DMA support issue that they have with RagePRO chips because of changes made to the XAA drivers in the 4.X XFree86 releases (They adjust the FIFO buffer size settings on the chip, which doesn't seem to impact normal MMIO operation, but totally breaks DMA operation- go figure...). I plan on conducting a few basic experiments to the current DRI code on certain chipsets to see if there's not a better way of handling the security of those chips. Also on the hit parade of things planned is some research into ways to provide accelerated indirect 3D rendering- preferably within the current XFree86 (or Xwin- depends on how that fork goes down actually...) framework if at all possible. There's chipsets that while absolute peak speed is possible with direct rendering, the securing of the pathway eats any performance gains with respect to the direct operation. There's also chipsets out there that don't have any DMA operation so they gain a lot less from direct rendering.

[Portions of this blog entry have been deleted for personal and business reasons- be CAREFUL of what you blog, as it can be embarassing or cause problems down the line... :) ]

I've done a little bit on the Ballistics port (Though not checked in yet... The team's making some pretty amazing strides- see LinuxGames for the first screenshot (BTW- we're beyond that by a long ways now...)) and I've thought a little more on the Tux Tag stuff, but most of my time is being devoted to packing up the house and putting most of the stuff in storage. Am I losing the house? Your guess is as good as mine. We've four weeks left before the official sale of the deed, my employer's about to get an emergency bridge loan from the VC that will re-establish payroll and insurance for us for long enough to count for the mortgage company (It's supposed to happen any time now, but you know how those things go...), and we've not heard the determination from the Loss Mitigation people at the mortgage company. I'm not going to wait and get blindsided- I'm packing up everything in anticipation of me having to give up or sell the house.

[This blog entry has been deleted for personal and business reasons- be CAREFUL of what you blog, as it can be embarassing or cause problems down the line... :) ]

eBay.

The few things sold so far have pretty much paid for all the other future and possible sales with a bunch of change to spare. Only dismaying turn of events was the absolute lack of activity with regards to the 10 3J Tech 33.6k PC-104 modems I put up- guess I need to sell those on a different venue... (Know of anyone that needs a small quantity of PC-104 0-70deg C modems? I've got more than just a few of them available for sale...) I'm expanding the sales with a handful of premium Allwell 1030 configurations and a bunch of baseline configurations. If you're looking for a good price on a known to be working Allwell 1030 for a set-top toy or some embedded/kiosk/etc. application, look for "svartalf" on eBay- there should be at least one up for sale for a while yet.

On the DRI front, it looks to be that the other two people working on things are going to end up coding a hybrid of what I was attempting on the RagePRO driver. If it is stable and secure, it'll be faster that what I had in mind- if it's not, what I had in mind will work very well and allow for 2D DMA acceleration, etc. I plan on continuing to help/work on the RagePRO driver, but I'm going to see what I can do with the Trident support angle (since I DO happen to have those chipsets in a few things and I happen to have a register spec for the CyberBlade in the VIA chipsets...)- I've contacted Trident to become a registered developer through my employer (What do you know, they ARE useful for some things after all...) and I'm moderately hopeful I'll get some useful access in the next couple of weeks.

Tux-Tag's still kind-of stalled, but I've just about settled on a design for the pack/gun IR transponders and when I've settled on the Mark-I design proper, I'll be putting that up on the website for download. An interesting twist in the whole thing is that I could just as easily as not use a Dimm-PC or something like it for the whole thing (which lowers the price even further than the MZ-104 board...). It may be just as well, ZF-Micro seems to have a problem with their suppliers (Seems that National Semiconductor's cutting them off- unless they find another fab, etc. ZF's probably toast (which is a shame...)).

On the Intelogis front, I've pushed the hack-job I did for the 2.4 kernel support out to the site. I'm looking at what all I'll need to add interrupt driven support (If you've got an ECP port, it'll work like a champ when I'm done...) with info in hand from some of the old Intelogis/Inari staff lending a helping hand. Once I get the interrupt stuff working, I'm going to see what is needed to do byte mode instead of nybble mode like it is right now for peak performance. I've gotten a dump of what's going on with the USB interface with the SystemLink devices, but I've not had time yet to go through and reverse engineer the protocol changes to the PLX layer to make usable drivers for the new devices. I hope to do so fairly shortly.

[This blog entry has been deleted for personal and business reasons- be CAREFUL of what you blog, as it can be embarassing or cause problems down the line... :) ]

Whoa... Was it January since the last log entry? Guess I need to put these up more often.

Well, I picked up yet another project (What wasn't too many already NOT enough?! :-) with the maintenance/extention of the Inari/Intelogis Passport drivers. For those that don't know, this was a device that was the first serious attempt to do powerline networking. The RCA SystemLink devices are a second generation attempt at the same (nice unit, no drivers save Windows, even though they're USB/Paraport devices...). I plan on modernizing the Passport drivers (I'm trying to network my in-laws out in the country and it does seem to work with the Windows drivers fine- so, I'm going to roll a demand-dial gateway and use the Passports to tie in the buildings with the gateway.). If I find time (Riiight... Time's the problem with all of this...) I will attempt to see what's different with the protocol with the SystemLinks and produce a driver for them.

On other fronts, in the DRI CVS there is a branch with my ersatz DMA manager code pending (I've got a few things to finish to suit myself before a check-in)- should be really good when I finish it. Tux-Tag's stalled for obvious reasons (I don't have time right now and I don't have the money in budget to build the gun/pack comm system, let alone anything else like the RF modem system I need as well. It IS going to be coming, it's just taking waaaay longer than it should. The VSBC-6 drivers have been languishing, but I plan on at least working the CVS over to make it 2.4.X compatible (No way for me to TEST this- if you happen to have one of these boards that you could "loan" me short or long term, it'd be appreciated!).

Oh, and there might be light at the end of the tunnel- I got most of the bulk of what was owed me in pay about a month ago and I'm only looking at being a month behind (I'm supposed to see another check by the 30th of this month.) The current situataion looks to be that my employer's likely to get funded and we're going to field something pretty cool or unbelieveably awesome (Depending on how they play what they've got in hand- I'm hoping for the Grand-Slam play myself...). This is not to say that I am not trying to find other solutions- if you know of a Senior Software Engineer position, I wouldn't mind you throwing the Info my way.

Ahh... Someone IS listening afterall... A friend from a previous job put me in touch with his employer as a possible candidate for future openings there- made my day. If you DO know of a job where they could use a developer of embedded systems, device drivers, etc. please send them my way.

On other topics, new year, a setback, some new progress. I had bits and peices of code in place for the DMA portions of the RagePRO DRI driver for a while now, but nothing concrete and usable- I'm finding a little more time now and I expect to be pushing a CVS update in the RagePRO branch shortly (Hopefully within the next week or so if things keep up the way they have been...). Also of note is the design specs becoming available for the Ronja IR wireless link. Either unit could well be a good modulation scheme, etc. for the IR subsystem for Tux-Tag and I'm evaluating the possible use of either the Metripolitan or Loopipe versions right now. In any case, that's moving forward again as well.

That'd be the progress...now, for the setback...

NBCI, formerly known as XOOM, for some inexplicable reason chose to discontinue their free hosting service (I understand this part) but didn't tell anyone about the change. While it wasn't much, I would like to have had the opportunity to change the content over (I had my online copy of my resume there amongst other things- it'd been nice for me to have had some warning...)- but, alas, they had other ideas for all of this. So, for right now (and maybe indefinitely) it's housed at svartalf.freeshell.org- not too bad once you either get ZMODEM working over telnet (I had to use Virtmodem under Linux- nice little application by the way...) or pay them the one-time fee of $38US for standard FTP, etc. access to the shell account. There is a $1 fee to authenticate the basic shell access (because some twit used the site to attack others with...) but that's okay- $1 for 100Mb isn't a bad deal and I plan on giving them more per year to ensure continued operation.

[This blog entry has been deleted for personal and business reasons- be CAREFUL of what you blog, as it can be embarassing or cause problems down the line... :) ]