Jump to content

Dr Memory

Members
  • Content Count

    103
  • Joined

  • Last visited

Posts posted by Dr Memory


  1. 1 hour ago, mimo said:

    Are you using the syscheck by tf_hh? Pretty sure that requires 64k base ram to work 

    Screenshot_2021-04-08-20-09-09-711_cn.wps.moffice_eng.jpg

    Hmmm.  I am, and I think you're correct.  I actually checked first, and saw where it said "Sys-Check also works fine with an Atari 600 XL", but somehow managed to miss the rest of the sentence.  Oopsie.  So that's probably a bogus bad memory report I'm getting.

     

    I'm still gonna have to socket the DRAM as a minimum, and probably a lot more than that if I'm gonna put in a video upgrade as well.

    • Like 1

  2. Wow!  Thanks for the memory upgrade info, guys!  I will almost certainly do that - I never intended to live with 16K.  :)

     

    I did a little more testing on this thing.  Three keys are dead - left shift, BRK, HELP.  Odd assortment.

     

    The memory passes the internal test, and SALT, just not Sys-Check.  I certainly trust Sys-Check more!  I was shocked to see all that red.  I have some known good DRAM chips and a memory tester so should be able to figure out what's going wrong.

     

    However, I hit a snag.  It took me quite some time to disassemble the thing, and I found to my surprise that it is not socketed!  I thought all 600XLs were socketed?  So I can't just pop the DRAM and test it, there will be some work involved.  Huh.

     

    440748555_600xlboard.thumb.jpg.fb425101b8977c7a7a3015e9ad4c207d.jpg

    So that's going to slow me down a bit.  On the bright side, it already has the video components filled in, which also seems unusual - I expected to have to get a kit or a bunch of discrete components and do that myself.   Maybe because it's PAL?  Are PAL 600XLs known to have their video stuff already populated and no sockets?

     

    The picture quality isn't bad, but is composite only, no Chroma nor Luminance.

     

    I'm letting it rest for a little bit here, had to use WD-40 to get a couple of the screws loosened.  It seems likely this thing was never opened before.  But if that's the case, then I'm very confused about the video situation.  I'm also very confused, almost shocked at the lack of sockets.  Guess I need to do some more research here.  And some desoldering, of course.

     


  3. Well.  This doesn't seem like a good thing.

    717263659_600xlmemory.jpg.69b799c1dfc171c97e1dbc8cae770c38.jpg

    It did say "untested"!  Interestingly, it boots, and I get a plausible response from ?FRE(0) (13326), but as you can see, it is not a happy camper.

     

    Keyboard issues too.  Not sure what else yet.  Oh well, opportunity to learn things!


  4. There was another one not too long ago that sold for over $400 also.  I don't really get it.  I bid on this one but they passed the max I was willing to bid like 2 days before the auction closed, and I was already feeling guilty considering spending as much as I was on such a thing.  Oh well.

     

    I'd like to get one someday because the very first Atari I bought, new and from a store, was a 1200XL.  I sold it because friends convinced me I really wanted one of the older models for compatibility with things.  It was probably the right decision at the time, but now I'd like to get a replacement for sentimental reasons.  The prices are just insane though, and frankly, it just isn't as important to me as it obviously is to others.

     

    Isn't that a good chunk of the cost of a 1088XLD w/daughterboards?  Sheesh! 

    • Like 1

  5. I replaced it with a #3 x 1/2 in Flat Head Zinc Wood Screw.  I think a #4 would have been all right as well, and I probably could have gotten away with 5/8 in but it would have been tight.  The head on the #3 fits in the recess nicely, so I gained a little depth from that.

     

    Cartridge now seems stable and much sturdier.  Still my favorite utility cart!

     

    I got a little frustrated in Home Depot when I got jostled and dropped (and thus lost) the original screw so went with the best match I had at that time.  Home Depot on Saturday afternoon is not the most mellow place to shop.  That's not really an Atari-related problem though.


  6. Did I make a pun?  Yes, I did.  If you expect me to apologize for that, well, I'm a frayed knot.  :)

     

    Anyway, I noticed that my SIDE3 cart was having a little trouble.  When I inserted it into a machine, it didn't always work right away.  I often had to remove and reinsert it and/or jiggle it.  So I took it out and fiddled with it a little and realized that it wasn't connected solidly.  In fact, the back pulled off with minimal force, and no screwdriver involved.  Huh.

     

    Looking at it more closely, I realized that the screw had eaten some of the plastic from the board side of the case, and as there wasn't much overlap between the screw and the case, it was enough to make the connection loose.  I tried to screw it in and it just turned...  That may have been an error to try but it was already hosed so I probably didn't make it worse.

     

    Looking more closely, I realized that the screw tapers off pretty quickly, and only penetrates into the hole on the board side a little bit.  Unfortunately, I dusted off both sides a bit just so I could see what was going on prior to realizing I should take pictures, but here is what I found anyway:

     

    Just the screw.  Note the abrupt taper.

    screw.jpg.4f3e0650c216172004e1a027c2d7a9ee.jpg

     

    Measurement of the screw.  The threaded part is about 1 cm, the tapered part maybe 2-3 mm.

    274420945_screwmeasurement.jpg.e948af4b8cbd0481db3189a8dc1e3a5e.jpg

    With the screw in the case, it only sticks out a little - that little bit is all it can use to attach to the other side.  Maybe 1-2 mm?

    1976956436_screwincasezoomed.jpg.df098e2e9236e9c6c778065074e26883.jpg

    Then on the other side, where the board is, the board sticks out over the plastic hole a bit.  So there is only a tiny bit of overlap between the part of the screw that reaches the hole and the hole.

     

    130503233_boardholesideangleview.jpg.064f393436a21407f423d4529e5b6206.jpg

    Seeing this, I guess it's no surprise that the screw wore away the plastic enough in normal use to basically fall out.  I guess if I can find a replacement screw with the same diameter but without the taper, I should be able to fix this pretty easily.

     

    I'll go to Home Depot later and see if I can find one.

     

    I love my SIDE3 cart and have no complaints, just this little mechanical challenge.  I posted in case others might have the same issue.

     

    If someone knows what size that screw is, it would be helpful to know that so I can skip the store search part of this.  :)

     

    If I screwed up the metric stuff, sorry, I was raised on the English system, but this stuff is so small that trying to explain it using fractional inches seemed like a bad idea.


  7. BEST is fine.  I've had pretty good luck ordering things from him and always get a personal response when I ask questions.  So he's getting older and may sometimes be a little crotchety?  Well, you know what, me too, to all of that!  :)

     

    I hit the 3-item limit as well on my last order, and really, it was fine, they were working off a backlog from the holiday rush and needed to control the flow a bit.  I'm sure if I go back now I can put in another order and get the rest of what I wanted.  You're not dealing with a big corporation, you're dealing with someone that does this as a labor of love.  So many of the people and companies I used to buy stuff from are no longer around and I'm grateful that he still is.

     

    A tip:  This is how I find stuff on his site.  In google, use "site:https://www.best-electronics-ca.com/ <whatever you're looking for>".  That causes google to search for the stuff, restricted to his site.  Works great for me!

    • Like 5

  8. 7 minutes ago, manterola said:

    Last batch it was  PLN350 for one ,or PLN640 for 2 pcs, including shipping. PLN350 is ~US$90

    OK, thanks for the info.  I'll stick with 3 for now, but would be interested in any quantity discount options if there are any.  I got the sense that ordering more may help because it lets him split the cost of the production run more ways, but maybe I misunderstood.


  9. I bought a keyboard from them on the 7th of this month (9 days ago now).  Meaning, ordered and paid for a keyboard.  No movement showed on ebay until yesterday, when it finally got a tracking number.  Same day they showed up as being out until the 18th.  I sent a note asking about this, and got a terse reply with nothing but the tracking number.  Grrrr.

     

    If I knew it was going to take this long, I would have done something different, and may yet.  At this point they are incommunicado and my keyboard is still not in the mail - tracking shows "pre-shipment information has been received".  I assume they did this to make it harder to cancel?  I've bought things from them in the past and had ok experiences but this is not a happy one.  If I'm paying $56 for a keyboard in mediocre condition I at least expect communication, if not the actual item!

     

    I bought a computer from a seller in Egypt a day later, and despite customs and COVID and weather issues and whatever else is going on, I received it yesterday.  It works great!  So this isn't an ebay issue nor a mail issue - both are working, in general.  In fact, I live in the same state as B&C, so buying from them ought to be quite fast, one would think.

     

    Obviously something bad going on over there in B&C-land.  Hope they're ok.

    • Like 1

  10. I have not yet done anything about the CRTs.  Saving that for a weekend when nothing else is planned.  Honestly, the video looks awesome on both, with the video circuits updated on the machines being used, they just have this quirk on the menus on two carts!

     

    I had planned to install my one and only UAV, which is a "Plug-In" version, in the system that already has RAMBO XL in it.  That system is pretty stable and I already fixed the memory problem, so it seems likely to get a lot of use.  Luckily did my homework first and found out that UAV wasn't going to fit unless I wanted to give up the shielding - that machine has one of the described "sort of wedge-shaped" top shields.  I don't really want to run RAMBO XL un-shielded, so decided to do the SV changes again.

     

    I thought for sure I'd remember all the details from the other two times I did this just days ago, but no...  So I took notes about everything I had to look up again.  They say memory is the second thing to go as you get older...  Anyway, I updated my cheat sheet and marked the locations of the relevant components on the 800XL board picture I've been annotating.  So, here they are.  As before, I take no credit for any of the design work done on any of this - I'm just trying to re-organize and summarize the info.

     

    Updated Cheat Sheet:

    image.png.897e278c3ec4bd942c902a9439e65cbc.png

    Annotated 800XL Board Picture:

    image.thumb.png.a55090a17caa75db8e13ea978ecee60e.png

    Relevant part of the board, the video circuit near U20 (the 4050 chip):

    image.png.f08e6c7f843ac626a88b85995773cb01.png

     

    I hope people find this stuff useful, or at least not too annoying.


  11. Oh heck, I have to tell the lightning story.  So I was still in the Atari scene, still the president of my local user's group and a well-known SysOp and even an ST developer, but I was also getting into PCs.  So there's the Atari link for the story.  *cough*

     

    The day came to upgrade the hard disk in my development PC and I sprung for the largest Hard Drive I had ever seen at that point that wasn't attached to a mainframe or minicomputer - a 100MB Rodime.  SCSI FTW!  If I remember right it cost me a cool $1200 bucks.

     

    It wasn't my first upgrade.  I was already building systems and setting up my own networks and such.  I powered everything down, grounded myself, added the new hard drive to the system (screwing it to the side rails on the drive cage, which was pretty standard back then).  I wired it up.  I still think I wired it correctly - Molex power connectors are certainly polarized and SCSI-1 usually had a notch or at least a red wire on the cable.  Attached power, turned it on - and ZAP!  Lightning!  Huge sparks!  Magic smoke!

     

    To this day I don't really know what I did wrong.  All I can figure is maybe Rodime drives didn't ground their cases.  It didn't come with documentation... Anyway, it sucked, and I had to eat it.  It didn't put me off computers but it did put me off Rodime.

    • Sad 1

  12. I have a couple of incidents that don't belong here because they don't involve Atari, though one does involve lightning!  Oh well.

     

    My biggest (or at least most memorable) Atari error was when I brought home my brand new 130XE from Fedco and tried to install an Omnimon/Omniview in it.  I probably learned about that thing from either a BBS or a magazine article - this was before the Internet really took off.  Anyway, I failed, and killed the XE.  :(

     

    So I ended up having to buy another 130XE (which was pretty stressful, I wasn't making much in those days and was living on my own for the first time) and pay someone to install the upgrade.  At least the chip was salvagable.

     

    I only got over my fear of going near a motherboard with a soldering iron very recently.  That took, what, 30 years?  35?

    • Sad 1

  13. 2 minutes ago, BillC said:

    The height of the shorter of the sockets in the link is about 1/2 that of a standard wipe socket, so using 2 stacked(1 soldered to the PCB and the other soldered to the IC) would have no significant height penalty.

    Ah, sorry, I misunderstood.  So that's actually a third approach!  Cool.


  14. Just today (yes, today) I was trying to debug a bad keyboard.  It has a dead row 3 - U, I, O, P, etc.  So after figuring that out, I took it out of the machine and did some physical inspection and realized I was going to have to disassemble it further.  I took off the 4 tiny screws holding in the function key assembly first.  *SPROING*

     

    I did not find all of the springs.  I did get it apart and still don't see what is wrong.  There are a couple of pretty corroded traces in the mylar cable but the one going to 3 looks fine, which I knew before disassembling things.  Oh well, maybe it's time to get a TKII or an AKI.  Trying to figure out which is better now.

     

    Anyway, I managed to get the keyboard out without breaking or warping anything, but still screwed up.

     

    I'll post a much older anecdote or two after I calm down from this.  Nobody to blame but myself - I should have researched the mechanics of the keyboard more thoroughly before touching it.

    • Sad 1

  15. 4 hours ago, BillC said:

    An alternative is to use a machine pin socket, insert the chip into the socket then solder a connection to any broken pins(wire may be required). This solves the issue of keeping the chip/pin stable and also reduces the risk of the replacement pin breaking off if removing the chip, just remove the chip/socket as a unit.

    I thought about doing something along those lines, either soldering the pin to the board or putting a stub in the socket and soldering to that, but I was trying to see if I could fix my POKEY in such a way that I might be able to swap it out again later if need be.  I guess you could still do it with your solution, you'd just have to un-solder the one connection to remove it.

     

    Agree that your way would be more stable.  I am not really that happy with the structural integrity of the two chips I fixed via graft (the POKEY and a Delay Line).  They do work though, so that's something at least.


  16. Coda - lotharek.pl is now accessible from where I am, as normal.  Cool.  No changes or exceptions were needed on this end.  So I'm thinking it probably was the CDN update propagation thing, but I'll never know for sure.  Such is the nature of the beast.

    • Like 1

  17. 6 hours ago, DrVenkman said:

    You can scrub that pin with Deoxit or other electrical cleaner, then solder on a leg snipped from a donor chip like a cheap logic chip or another dead or damaged IC you might be able to find.

    Just wanted to confirm this approach.  I managed to snap a pin off of a POKEY while I was practicing desoldering/socketing as part of my current retro Atari computer binge, and was indeed able to graft on a lead.

     

    Tip One: I've been hanging on to leads from resistors and capacitors and such as I snip them off, if they are long enough to seem worth it.  They are pretty handy for sticking in a breadboard hole to bring out a signal so you can clip a probe or whatever to it.  Anyway, many of them are really thin and wimpy, especially in modern resistors.  But, maybe 1/4 of the lead fragments I had stashed were noticeably thicker and suitable for pin replacement surgery.  You can always snip one off yourself from an easily replaceable discrete component if you don't happen to have a suitable wire fragment laying around.  I found posts that recommend using 30 AWG solid core wire for this, but I can't confirm that.  Anything that seems solid enough to hang in there and about the same diameter as a pin should be fine.

     

    Tip Two: Don't trim the lead fragment or wire or what have you to fit until after soldering is done and satisfactory.  It's hard enough to keep the IC and wire stable without also trying to manage a 1/8" or 1/16" piece of wire and a hot soldering iron!  Also easier to see the length you need after it's on and you can just eyeball it against the intact pins.  I also wouldn't recommend using a leg from a donor chip unless you can cut it a little longer than is strictly needed, so you have a bit more to hang on to.

     

    Tip Three: You need to hold the pin and the IC very stable to do this.  So if you don't have 4-5 hands, it helps to have something like Helping Hands or the like.  Even with that, things tend to skitter away when you touch them with the iron or the solder.  It can definitely be done, I've done it twice now, but it wasn't fun either time.  Really the actual tip here is, please be careful not to burn yourself!

     

    If you already knew all of this or it was obvious, sorry.  I'm just very excited at learning how to do all this sort of stuff, so trying to help where it seems like maybe I can.  Lord knows I've gotten a lot of help from the people here.

    • Like 1

  18. I tried to find an answer to this, as there are many keyboard issue threads, but was unable to.  If this has already been solved, please direct me to the correct thread, or at least something resembling the title of said thread.

     

    I have an 800XL whose BREAK key works in programs, such as the REV. A version of SUPER SALT.  It also works in BASIC, in the sense that it makes a new line happen and aborts the current line.

     

    It does not work in the built-in self-test.  What the heck?  How can that even be?

     

    I tried swapping the 4051s and cleaning the ribbon connector but neither had any effect.  And honestly, this was already a very clean connector.  It's a Type 5 keyboard and seems to be in really good condition except for this problem.

     

    I found several threads where people had the opposite problem, where only BREAK worked, but none where BREAK worked in user programs but not in the internal self test program.

     

    This is the same machine with U1MB installed, but that seems to be working great, and I couldn't find any menu items that said anything like "make BREAK key act funny".  :)

     

     


  19. I slept on it and thought of something else to try.  I got on a Linux box I use for development and tried to go to lotharek.pl.  Same problem, exactly.  Same error and still no opportunity to add an exception.  I then brought up a VPN and temporarily placed myself in the Ukraine.  This sort of thing is exactly why I subscribe to a VPN - I actually do network development and debugging as part of my day job on occasion.  Which is why I know how to debug things like this and what evidence to gather and how to think about it properly.

     

    Anyway, from there, with the VPN up, I can access lotharek.pl.  Same computer, same network connection, same web browser, same settings, no exception required, just accessing it from a different geographic region.  I think this puts the theories about this being my personal setup to bed pretty thoroughly.

     

    Best guess - Lotharek's website provider uses a CDN (Content Distribution Network) and that CDN is being slow about updating some of their mirrors.  The fact that I can access it without issue when localized elsewhere means his config is fixed, it's just that the fixed version hasn't propagated to my area yet.

     

    To be clear, if this is really an issue with the CDN and not with his site, he doesn't need to do anything and it should clear up on its own when the CDN gets around to updating their mirrors.  From our point of view, it should just start working.  However, if it doesn't start working after a day or two, Lotharek needs to open a ticket with his hosting service and tell them about it, so they can in turn open a ticket with the CDN.

     

    If, on the other hand, he's totally self-hosted,  and subscribing to a CDN directly, he has some work to do.  :)

    • Like 4

  20. 1 minute ago, Stephen said:

    People - DNS entries and SSL certificates don't just automatically update for all 7 billion people at the same time.  Calm down, it will all resolve in a day or so.

    Agreed - it is almost certainly fixed or in work and will eventually propagate.  I just got annoyed by some of the claims being made, sorry.

×
×
  • Create New...