Finding an intermittent fault on the Book8088 & adding external CGA output

well hello everyone and welcome back to adrianne's digital basement on today's video we're going to be looking at something old but it's also something new and it's something broken it's a book 888 laptop and this was sent into the mail call by viewer Ezra this was a little while ago there's the machine if you're not familiar with the book 888 it is a replica 888 laptop so it's old and then it's using an 888 processor so this is an XT class machine but of course it's new because this is modernly built and it's definitely broken so in this video I want to see if I can get this thing working again and because I'm recording this before I've even looked at this thing this video might be a complete failure so without further Ado let's get right to [Music] it I'll put a link down in the description to the original mail call video where I open this thing up because I did some rudimentary testing in that video and what's in this little bag here are at least some of the extra parts from this machine I think uh this thing has an opl3 option which goes right down here it's a little sound card with an actual Yamaha op3 chip on it so this is that little sound card there and it has a little board to board interconnect right there and I think uh four little screws which looks like I put them back in the standoffs yay for doing something right for once we also have a USB port and a couple LEDs which I think are status LEDs of some kind on this side a power switch looks like a audio output Jack for the sound and on the back of the machine we have a DC power input and then we also have a bus connector and this is actually Isa bus and there are actually some modules you can plug into the thing to add some expansion cards now the case as you can see is a little bit translucent we can actually see the inside we have a lithium ion battery and under this little compartment here if we pop this off I think we have a couple ROM chips here one for the CGA card one for the internal system itself and then right here we actually have the crtc the CRT controller that is used for the CGA card now this machine when you power this thing up does appear to be completely dead or I think it kind of powers up but doesn't really do anything after that um let's just turn it on here see if anything okay we got some beeping out of it um okay well this is a really boring video because it just works now wait a second um okay well I am not kidding that I was done with this thing with the mail call episode it did not work I put it away in the original box and that was that um I don't even know where to think with this so it looks like these chips are here I think Ezra sent these there was a new CRT controller or actually this is NEC part here so we have a new v20 processor here then we have a new SRAM chip this is probably for the CGA and then this is the original ROM bias which I went ahead and installed a new one all right we're looking at something that's like super meta here it's me watching my own video with myself it's like I'm I'm commentating on it or something all we can see in this original video is that what you happened when I turned this thing on is I think it did nothing there was no beeping or anything and the screen would light up and I would say something like no signal on it and yeah it just sort of appeared to be dead we got some flashing lights over there on the side I wonder if those lights are doing anything right now yeah with the system actually powered up and working we're just getting this back and forth flashing there and I assume that is for the battery if we look there there's like a picture of a battery it's not plugged into the wall right now hence the lack of charging LED but we can see very clearly we have a working system this is completely unexpected I really have I just can't believe that this thing is actually working now the glare though the glare and the screen this is a terrible screen the glare is pretty bad it's hard to record this and there's no video output on this machine that's one of its uh failings I guess so to speak all right so back to the video is there anything else I need to remember about what I did here I had it plugged in definitely when I powered it up it' just say here I think I'm going to do it in a second here I have the audio turned down obviously oh look it's me talking oh and there's Ramy all right yes very meta myself and myself that's a that's a younger version of myself this video is from two whole months ago okay so here we go we turn it on you would get this no signal that was that so it's like the screen module seems to have some kind of a I don't know must go from CGA I'm sure this thing outp puts a real CGA signal it goes from CGA to whatever the LCD needs to be driven with so it's kind of like it has an RGB to HDMI in it and I think what I want to do actually today since this video is I don't know not going to be a repair video uh let's re install everything on this machine like the opl cart and I'm going to still open it up because I want to see if I can find those TTL RGB signals and actually output them maybe from some wires that are sticking out the side of the machine so we can actually drive an external monitor with TTL RGB it's just 5volt TTL Logic for sync and all the video signals so there's no reason why you couldn't actually have it running the internal screen and an external monitor at the same time typically the output drivers are sufficient enough to be able to actually drive both of those things and I've done that before with real CGA cards and it to does work now just to reiterate on this if we take a look at the little trapo area here so 8057 is not installed there's a new v20 that I installed so I took an extra one and stuck it in there and I also put this additional bios chip that I flashed off of code that I found on GitHub and after doing that there was no change in this system I did that in the mail call episode and the system absolutely did not work so between the two months that I had the sitting in the box and now this thing just magically fixed itself part of me has me wondering if there's just a loose connection so let's just turn this off and let's bang this around a little bit and let's see if that has any negative effect on it yep it's broken again all right so at least the problem is back and this is exactly what it does the screen just kind of does this refreshing thing and we did not get that beep anymore so interesting let's um let's bang it this way and see if that makes any difference power this on there now we're getting that interesting so this does seem like just a loose connection of some type inside of this thing and now we're getting nothing again now I'd already receded all these chips in the mail call episode and that didn't seem to have any effect there we go it's freaking working again oh boy good quality there great quality yeah all right let's take a look at this bios while it's actually working I have to actually look through the camera view because I can't read this off angle at all right now anyhow we see we have a v20 processor which of course makes sense with what we have installed in here we're running in CGA 80 x25 we do not have a math code processor installed there are no comp ports on this machine which is totally expected no parallel ports for some reason it says we have floppy drives maybe that has to do with the USB port that's on the side maybe it emulates a floppy drive we have 64 40K of conventional memory and then bias ROM extensions are at f0000 and I think that is uh basically the XT IDE and of course there's the disc is not installed I have the compact flash card removed and this is the normal XT IDE stuff here at the top let's hit control alt delete I don't think the keyboard is actually working so even though um this thing is working I'm hitting control delete and nothing is happening so this was included with the machine from Ezra so let's try booting this thing up see if this thing with a power cycle at least power it off power it on is it actually able to see that card all right testing Ram let's see if I hit Escape oh okay the keyboard did work oh it's got flicker like it's got snow there so it does see the flash card and it's whoa why is it so slow look at how it's updating the uh when you scroll that is far worse than that what's happening that is not normal for um a CGA it should not I mean you're going to get flicker but it shouldn't be that bad what I want to do is get this card out wow you really have to use your fingernails to get that out and I'm going to put my card in there my card is uh well it's just say it's filled with all the usual utilities that I do testing with on the channel now from my understanding the CGA circuit tree that's on here is a replica of the original CGA card and I think it's somewhat compatible hence the um oh yeah no high mempis this is not configured for an 888 let's reboot and I'm going to hold down the uh the shift key for whatever reason there's aom key right here it just shows where the keyboard is from that is hilarious I do have to wonder do these Keys work I can't imagine they do and is the computer Frozen because it should have exited out of this stage already okay I power cycled the machine holding down shift that actually worked the performance again of the BIOS routines are unbelievably bad now I'm wondering if the bad performance on this thing right now has something to do with the open source bios that I actually have installed on here it could well be that that is what's causing that because CGA itself does not have its own bios that extra CGA chip we saw on the bottom is actually something like a character ROM all right so I just edited the auto exac and the config sis let's reboot the machine here and now I should boot normally that beep sounds a little bit loud and annoying uh looks like on the reboot it doesn't synchronize the screen properly and like cuts off uh right oh I didn't take everything out that should have been out let's take a look at top bench from Jim Leonard let's see if this thing is actually performing at the speed I would expect this thing to be running at I don't know if this runs at 4 4.77 MHz or this actually runs at say 8 MHz like a turbo XT now this horribly slow performance might well be what you get with CGA on top bench for instance running on a 4.77 MHz machine but I don't think so and I know for sure typing dir on a CGA 4.77 MHz 5150 is far faster than the way this thing looks performance-wise we're as good as an IBM PC junr running to 4.77 MHz with a v20 and those results actually ones we're looking at right there that we're comparing this thing to were submitted by Jim himself now interesting is we take a look at the mem test a little bit faster on this machine meem EA is about the same and op Cod is about the same so absolutely this machine seems to be forming at regular 4.77 MHz speeds and I just don't understand why the bias routines are so slow when you're doing a dir command that's I guess the only thing that's out of the ordinary I guess I'm just not used to seeing top bench load on a real 4.77 MHz machine and see it going that slowly now I did a little quick Google because I don't even have a manual for this thing and I really don't know anything about the book 888 but I found this thread here on VCF talking about some function keys that actually do work so first one is function one for mute that doesn't do anything at all on this machine function two sound on function two sound off nope I mean there's still a beeping or a clicking coming out of the machine the only thing I'm thinking is maybe those sound on sound off thing only has to do with this opl2 card all right the next one that's kind of interesting is function six turbo mode mode with no indication of mode Let's press that function F6 oh that works this machine now gets a score of nine which is the same as an overclocked IBM PC Jr running at 8 MHz the PC junor 8 MHz speed are those scores and these top ones here are this machine and you can see they're all a little bit faster on this machine than they are on the PC junr so probably means that the clock speed is just slightly faster on this now in case you're wondering what that flickering is that is not a fault that is actually the original CGA adapter from IBM did not use dual ported memory or any strategies to avoid corruption if the CPU was updating the video memory while it was being drawn to screen and that's what's going on right here is that the replica CGA card that's in this thing right there is actually replicating this original card which has the effect of causing this snow now one way the IBM mitigated that snow in regular dos programs is whenever scrolling was happening in DOS like if you type dir whenever the scroll happened which was when the screen had its video memory being updated it would actually blank the video image at that time and then it would come back and it would have the next frame of video so has the effect when you type dir as every time the screen Scrolls which is when it updates it actually goes away for a split second and comes back it's a little bit of an annoying thing and later CGA cards like that awesome VTEC card that I showed off on the second Channel recently did away with that problem and they Ed two different strategies for making that happen they either use dual ported video memory where the video memory could be accessed through one set of ports for drawing to the screen and the CPU could access it through the other set or it just used a strategy to make sure that anytime the video card was drawing the screen it would not allow the CPU to read or write from the video memory and the CPU could only access the video memory during the vertical or horizontal blanking intervals which are around the edges of the active picture now looking back at this post here it seems like the rest of these functions aren't really particularly useful unless you want to use these extra keys and uh mode 80 is just like I guess puts the computer in 80 column mode if it's stuck in 40 column mode the post is also turn about the Sergey bios that is actually what I ended up putting on this so yeah this is the GitHub bios right here that is what's on this machine now I think the original bios chip that was put on here used a lot of this code or all of this code without any attributions so it would be a license violation specifically all right back on the thread looks like we have a few other mods for the machine so there's a hack to add Dam refresh Cycles back to the hardware which makes it far more accurate I guess running at 4.77 MHz it's like Freddy who's also the person working on the awesome Pico m project I did talk about that very briefly on one of my mail call videos as well has an improved driver for the USB port that's pretty cool looks like Joshua did some work on getting a USB mouse to work looks like there's a resistor mod to make the screen brighter and it looks like on board there is a 10 pin header that is for CGA output it mentions schematics for this thing so I guess we need to go find those because I still do need to figure out why this thing seems to fail working as soon as I bang it around a little bit it I found a post on RS Technica from Andrew Cunningham about this machine and there are some tear down pictures and it looks like this thing is absolutely just using single board construction now it does have these large ic's and they are in sockets at least on this picture here and I'm wondering if the problem is is that some of these are just loose and that might be why banging the computer around seems to change its functionality so it might be as easy as simply pushing these back in now with this machine having been on for a little bit of time there's definitely some warmth right here over these large chips so let's just pop out the nine screws that this thing looks like it has on the bottom cover to hold it down all right so appears that the screws don't let the bottom cover come off actually seems to remove the top cover here how do this All Connect up all right so there's a thin membrane here for the keyboard typically with these little things you have to either flip it up there we go and that comes out going to say the other possibility sometimes it slides that way and then you can slide the ribbon out we have a cable here which I'm assuming is like a power cable for the inverter board or whatever the driver board that's in there and then here is the driver for the LCD so we'll just connect that and then there is another little cable right here which maybe is like a backlight power cable or something once the top is removed looks like we have another set of screws to remove here that takes the PCB off of the bottom frame and looks like the battery has some adhesive that holds it onto the bottom cover or is it the top cover no it's the bottom cover so we're just going to use this plastic spudger here to disconnect the battery cable there's the main board it's exactly how it looks in the photographs so there's that battery that's uh left behind and looking at this I think what we see here is a lot of the logic consolidation into these large ic's that really makes this thing possible so if you look at the original 5150 motherboard 5160 motherboard there's a whole plethora of chips Intel chips and stuff like that and of course on later XT motherboards that was all condensed down to chipset ic's which were soldered onto the motherboards and greatly simplified things problem is those aren't really available anymore maybe you can find some New Old Stock or ones that were removed from boards but when it comes to manufacturing something new well I'm not quite sure what they did here did they actually recreate that stuff like here in this xdo chip and looking here in this area we have some system memory we also have this chip right here which it says there is the dma controller the 8237 so that is still available as a discret IC compact flash card is at iio address 300 and there is a USB controller which is this IC right here which is found at iio address 260 of course that's not something that's like normally found on on an XT but you could easily connect one of these to any XT and it's just it's available on a regular IO Port these are bus transceivers right here 245s these are 573s which I think off the top of my head are latches and I think I recall seeing somewhere that obviously the creator of this project took these two ic's along with the crtc the font ROM here and the SRAM for the video memory and put that on an Isa card for a new modern CGA card which uses this circuitry right here now when it comes to adding CGA to this system here there's that expansion header j6 that was talked about in that post and I am positive that that is exactly what this is this is CGA and if we flip this over here and we take a look at what's going on on the board here you can see here there are noninstallment ship looks like you see that right there and look how this one here is right up at the Top If I give this thing a good push like that it is now properly seated in that socket just like this I see I think at some point this thing must have got bumped and the chip sort of popped out now that one there is actually in all the way and it actually sits well a little bit proud of the socket itself it's a lowprofile socket but these ones here they needed a good push okay I just reconnected the battery up all right so we have power LEDs but we don't have any beep this thing is not working again so I'm just going to tweak the board a little bit okay so it's working now if I tweak it the other way well now it's working so what's going on we have we have some kind of bad connection on here clearly this looks like rework right here on this Ram chip so this thing has had some repair work already done to it looking at the way these are soldered to the board this all looks like it's machine done but you see how messy this looks right here my only assumption is there's probably some issue going on right here still so let's just push on this chip right [Music] here I just want to see if like you know bending this board results in this thing not working so far it's okay but this machine absolutely was not working when I first plugged in the battery just a moment ago so we're still not out of the water entirely here there we go it's dead again so what exactly is causing this let's push on things all right so the io chip maybe right so it's dead I'll push on the chip still dead let's give it a harder push well I'm going to pop out this IO chip here let's get this out of the board without this obviously the system is not going to work at all like I can turn this on and we're going to get that dead Behavior y so what are we dealing with here are we dealing with a bad connection on one of these pins these do look a little bent interesting why why are these not just perfect I assume this thing is a new chip and not some kind of New Old Stock yeah these pins are in kind of rough shape there's like a little bit of something down there a little something something I don't see any corrosion on these legs they just are a little flattened and slightly mangled so one thing you can do with these plcc chips is you can use like a pick or something you can actually bend them out which just makes them contact the socket a little bit better and then of course I mean Adrian's digital basement here right I'm going to put deoxide on this okay so all of those chip legs have been pushed out a little bit so they should make slightly better contact with this low profile socket now unfortunately this is surface out there's really no way for me to check to see if those legs are all connected but I'm assuming I mean there could be a problem with this socket here and that that might also be the issue here but I'm just going to put a little bit of deoxit on all these pins just to help things out all right you give this a firm push down and let's see did I make it worse or is it the same it's not not working now at all so yeah the system is on the lights are going there we have no one home really feels like it's in this area here doesn't it pushing on the PCB itself that didn't do anything we push on this IC interesting it starts working let go not working push on this IC okay no change all right this IC right here it really screams that that might well be the problem let's just get this out as you can see that we have an Harris IC from 1995 this IC might just be flaky luckily there's nothing custom about this chip so it would be totally doable to find a replacement for one of these ic's I could just steal it off another board as a temporary test as well let's just put some dox it in that socket there make sure that's all the way in turn it on nope let's push down no system is not working now could well be a bad socket though let's push down on this IC again yeah so it seems like I've made it [Laughter] worse let try to get it all in the view here just you can see what I'm doing so tur it on all right all right yeah it's got to be this Harris chip maybe I should just Reflow these solder connections they look okay I don't see anything that looks bad but can't hurt to just Reflow them right so I'm just heating up each pin till it moves and there's plenty of fresh solder on the tip so if anything's missing from one of these it will will get Wicked in okay there we go let's pop this back in do we have a working system we do now I got a bang on stuff here and it's dead again all right here's the data sheet for the renais version of this chip it's a seos bus controller I think this generates some of the ISA signals that are used on Isa bus this works with various 888 8886 and well early processors all the way up to 8 MHz here's the block diagram for this chip it takes some uh signals that come from the CPU there takes a clock an aen which is also from the CPU you various control inputs and then it outputs these logic signals now it says here multibus command signals is this compatible with Isa bus I'm assuming it is because this thing has a compatible Isa bus uh and it has these other signals here that go and drive the data transceivers that's the 28 45s and things like that now with the system sitting here on the bench it's not working very reliably right now it just doesn't seem to start up at all if I pull it towards me the system pretty much works if I if I push it away from me it does not work I'm really wondering if this is just a bad socket I know I know I refl it but this really does seem to be specific to this chip and I say that because I'm just going to pull over on the chip in the socket here and the system is running there and if I let go it doesn't work so I didn't I was careful not to apply any pressure to the PCB itself or any of the other ic's in the area I'm just going to basically squeeze between the chip and the socket there oh and look at that I mean think that pretty much confirms it's either the chip that's bad or it's the socket that's bad all right so I just popped the chip out of the board ever so slightly and I'm just going to push it in partially there we go it's not running I'm going to push on the chip very very interesting now everything looks okay on the chip I mean the legs are a bit bent up so this obviously came out of something else so we'll just straighten these out a little oh one of the legs just came right off the little leg it's over here on my bench I did not bend that I did not bend it over that must have just barely been hanging on I'd say this is one of the issues with chips that are pulls from things because you just don't know if someone's gone ahead and you know rest straighten those legs and you end up with ones that are just ready to absolutely fall off and you can see the one of these is longer than the other this one so I don't know what's going on there all is not lost I'm just going to solder a new leg onto that broken pin all right I have a dead eom here I'm just going to borrow one of the pins borrow steal one of the pins off this bend it around until it metal fatigues right off then you end up with the pin so we'll just tin what's on the IC itself and then I'll tin the the leg itself the one we're going to be Hing on it's attached to the iron right now that is fine and then we'll just get the connected together doesn't have to be perfect just has to be attached all right there we have it just going to trim it off so it's exactly the same length as the rest and while it's kind of ugly that should do the trick and will it work no so that was not the problem let's just move this again so that pin that fell off was absolutely not our issue I think what I'm going to do next is I'm just going to place this socket there are the reminisces of the socket there's a new socket installed in the board it's uh one of those machine type sockets I like those a bit better I know there are lots of people who don't but I do so that's what's in here so have I made this worse it seems like I have well at this point system appears to be totally dead in the water all right I think I'm going to have to break out another system so this thing is totally dead let's just try one more time yep absolutely nothing happening now well I have another motherboard here this is an XT clone motherboard and it actually has that same IC right here next to the math code processor if we turn this thing on ah well it's not working this machine worked literally a moment ago and I can tell because uh we're not getting any post codes on here we had post codes and I had a beep and everything was hunky dory and now the system is dead I'm not batting a thousand here I am really really striking out on this system why did this system work literally a moment ago and now it does not work all right well anyhow the chip we're looking for is right here on the motherboard I'm just going to take this out and we're going to pop this into the other motherboard into the uh book 8088 what I was hoping to do is actually take this uh 8288 and stick this in the other motherboard which was working a moment ago and that would be a perfect test oh the one I'm putting in here by the way it's a UMC part okay the UMC part is in and it freaking booted up it freaking booted up that means that this Harris chip I'm assuming is going to be the faulty part let's just give this the little push on this area here yeah unbelievable I guess there's just something flaky about this chip on the inside perhaps I wish the other motherboard worked right before I started recording it worked I I promise you it worked and I was I was intending to swap these and then take this Harris part from here and put it in the other motherboard and see if that did the trick I was just about to put the XT motherboard away and it's running now so the postcard has a code of A8 on it now I XTS don't really have post codes but if I pop a video card in here this machine will actually do something usually oh no it froze again yeah it says FZ yeah okay I have the chip in the motherboard that actually came from this and that's what was in there just a second ago when this thing actually did boot what happens is you should start to see random post codes on here when the motherboard is working and it is not doing that it's coming out of reset now at least no now it's stuck in reset there just seems to be another flaky chip on this thing and I don't know what it is now I did put the v20 the one that came with the book 888 into here maybe that's a flaky part as well I'm just just just for sure I'm just going to remove this and we'll pop in a known good processor in here and not this one that I wasn't able to test before here is a nice Intel 8 88 I think this is a 4.77 MHz part and that's fine this is not a turbo motherboard let's just pop this in here actual Intel part here okay it's posting so let's plug in the video card here let's see maybe that ship was flake as well uh yes yes okay you got to be kidding so this v20 might also be a dodgy part as well I thought this was dodgy in the mail call episode I thought this looked rebadged that is not what's in the book 888 it has a real 88 in there now let's pop out the 8288 so this is the UMC part that was on this motherboard and it was actually soldered into there and we'll take out this Harris part that was on the book motherboard and I'll stick this into the motherboard here and absolutely it is uh not posting there you see the two dashes on the postcard let's power cycle it again same thing just two dashes I'll put back in the UMC 8288 there there we go all right well that pretty much confirms that this chip here is absolutely dodgy and and my assumption is when I was pushing on it that was somehow making it work maybe there's like a bad Bond inside one of the the legs that goes into the body of the chip who knows but it seems to be completely dead now well and we just pretty much confirmed it on this motherboard all right so the next question is is this CPU here this v20 that came in there also bad let's remove this 888 which does appear to be fully functional it's quite warm that's normal and we'll put back in this sketchy absolutely repainted v20 here into the motherboard it did post so I know this chip does sort of work it's not totally dead but it appears to be flaky as well so into the motherboard it goes and yep not posting same thing we're just getting dashes wow just for the final confirmation I'm just going to put the the true genuine Intel 888 back in the motherboard here power this up there it is posting all right well I'm going to say that this chip is also bad so two bad chips in the book 8088 so that means that that socket I took out of here probably wasn't bad but that's okay I don't mind putting anyone of these machine sockets I just find them a little bit better anyways so the UMC 82 c88 is in the motherboard and we turn it on and it just works and let's bang on it bend it do stuff to it yeah okay so even though these weren't really pushed in all the way that could have been an issue I don't know but pushing them in all the way didn't really help anything it was a red herring that the 82 c88 was actually just a flaky chip I thought the socket was bad cuz you saw me pushing on this and it would sort of work but no the chip itself was actually bad and incidentally the chip that's in here now is from 1986 16th week well the diagnosis that I did here was a little bit backwards I like to kind of try to see the problem but because everything was in sockets it was really easy for me to sort of push around take things in and out of the board and whatnot and that definitely brought me to that chip as being the flaky one now on the other hand I think it was kind of Lucky this chip started working because if it wouldn't have started working it would have been much harder to figure out what was wrong on this because at that point you're just blindly going around the board and you're checking for signals in and out to try to see if you can see missing signals that are keeping the system from running so now I know this is the bad ship let's scope all the signals on The Good Ship while the system is running and then let's put this into the board and see if we can try to identify Maybe signal that's not working to help me more easily probe all 20 pins on this because doing it on camera is a little bit difficult with the camera over my head I'm going to use one of these chip Clips so this just Clips on top of the IC itself and then it gives me these pins on the top it just makes it much easier with this on here for me to get to these backside pins because I'm looking at the board this way right with the camera overhead and that way I can just uh very easily probe these and we'll start at Pin one which is just low and that's what we saw before and then that's the clock signal pin two pin three is lots of activity that's the S1 signal pin four is the DT SLR signal so we see plenty of activity there Al is pin five we got activity aen don't see any immediate activity but let's power cycle the computer no don't really see anything there but maybe that's just normal for this particular computer pin 7 has activity py cycling the board just stuff happening pin eight plenty of activity PL nine we got activity as well and then 10 is ground pin 11 which is IWC if I zoom in and out here we can just see it has this pulse whoops I lost the trigger point there so we zoom in there it is all right check this out check out the width of the pulse there on pin 11 Pin 12 it's a little bit wider pin 13 is a little wider yet pin 14 we see two uh and what's this this is pin 15 Cen this looks like it's TR state see how it's sort of floating in the middle there so that just must be normal for that particular pin here we are on pin 16 Deen plenty of activity here's pin 17 the MCE pin then we're on S2 pin 18 pin 19 is s0 so we got activity there and then we have VCC so that's 5 volts I'll swap over to the non-functional chip the issue is here is that there's a lot of complexity going on with the Intel type system like this and I don't really fully understand the way it's all supposed to work so like what are the input signals what are the output signals what signal missing from this chip causes this to not boot so the Chip's in there let's just make sure it doesn't boot yep okay we're still dead and now I wish I could remember how everything looked pin one doesn't have anything which is expected that's the clock which we know we have good so S1 this signal here had stuff on it and there is a little bit of low there for a second and then it just stops oh okay we still didn't have a booting system we didn't hear that chime I'm on pin four now we power cyle the computer we have stuff happening but the computer's not booting all right so this is pin five Al I could have swore we had activity on this pin and that does not look right now we have to see if Al is an input or an output if we scroll down here Al is an output signal now this could be our problem right here that's the bus driver on that particular pin on Al let's just double check by swapping back the other chip pin five if I turn it off we should have activity so you saw it was sort of there was what looked kind of like activity there but not quite right it wasn't really going down properly the working chip is back in system is running so one 1 2 3 4 5 okay and I'm on pin five and yeah we have a good signal here and this is an output signal so right there that's kind of telling us that we have a bad output signal on that chip now one thing we can do here is I can pull out this chip and let's look at what that pin looks like on the motherboard with that chip not installed one two three four five okay okay I'm on pin five and we turn it on okay so we're not really getting anything it does float High we're at about five volts so that's normal let's check out pin five now well pin five looks good actually one two 3 four five oh no pin five doesn't look good oh it kind of comes and goes when I push on it it kind of comes and goes and if I push on the chip on towards the body of the chip it does that so let's hold the I'm going to hold the Probe on it and we're going to power cycle the computer and see if it runs now oh man that is it right there everyone we can actually see the failure so it's failed right now and if I push on the chip on the leg itself there it is it starts working and if I power cycle the computer uh no the signal is bad still I got to push push up towards the top there we go I'll push I'll turn it turn it off and on yes that's it so confirmed we can see the fault there must be something going on now this is the good chip where the leg goes into the body like some corrosion or whatever and pushing on it like physically pushing on it with the probe actually makes it connect and we'd see the good signal but as soon as I let go and I touch it very lightly or I touch it near the bottom we'd see that kind of signal that's not really there and with the pressure on it the system working as soon as I pull the probe away the system instantly crashes so we've kind of confirmed right there that that is The Fault With This chip I mean that's one of the faults who knows what else is wrong with it but this chip is dodgy ass so that's good I'm glad to know that we can actually see the fault on here now of course if you were just probing around on this system and you didn't know what it should look like it would be difficult to know that that was a bad signal like we know it was bad because I could see it changing or when we had the good chip in there what it looked like but I have to admit it didn't didn't look bad enough it was sort of just there was a little bit of signal up at the top that would kind of give me a clue that it was trying to drive that signal down to ground but honestly you know TTL logic is noisy and something that's up near 5 Vols if there's a bit of noise up there that that's often totally normal I mean you look around the commer 64 those logic chips these are camos chips actually so seos drives the bus much harder than regular old TTL logic you look at a 64 and it's just noisy all over the place so it's very difficult to look at signals and if they're at 5 volts or at Zer volts it's hard to know if that's a fault or not unless you know for sure that chip should be toggling that output for instance based on the inputs when you see something on a 64 that's at like 1.2 volts or it sort of goes up to 1.2 and then down to ground but never goes above 1 .2 well yeah that's a problem or vice versa if you see something like up near 5 volts and it's kind of pulling down to 3 volts to go towards ground but it never gets all the way to ground stuff like that can be a little bit more of an indicator of an issue but this one that's a that's a really tough problem to solve now I'm glad it was something that I happened to have a spare so I was able to test this thing and yep I have the working one in there and it's just running now and if this was a problem on one of these customizes on here well there would be absolutely no way to fix it and it would be very very difficult to troubleshoot all right so we know the system does appear to be working we have two flaky chips the CPU which um yeah I guess it's kind of flaky as we saw on the other board and this uh Coss bus driver here so next let's try to investigate the CGA output and see if we can get that working because with a CGA video output I could bring an actual 1084 computer monitor and bring this up to one of the computer meetups I might go to periodically and have a little portable XT machine you'd have to use the keyboard on here you can't plug an external keyboard in but nonetheless that is still kind of cool and with a real monitor you're not stuck looking at that frankly awful LCD screen we have to figure out what these pins go to so we know that there are these resistors that aren't installed here and this makes its way over here to this section here which goes to the LCD screen and there's a little piece of metal on my my thing there when you look at these resistors installed here in these dials this appears to be a voltage divider I wouldn't be surprised if the LCD screen takes analog RGB so 0 to7 Volts for RGB and you notice there's just RGB there's only three signals that go here plus these two at the top these are probably sync signals horizontal vertical sync and this is converting the digital TTL signal which I assume is on this board into something that's analog I bet you what this is not doing is that the dark yellow color is actually not showing up as CGA Brown as what it should do with the IBM CGA standard is probably just a light yellow and a dark yellow through here I have to check that later but if we move over here I put a dot right here on what I think is pin one and I just looked at this connector here for the ISA bus and it has the same uh pin one right there and the way these headers go when it comes to pin out as opposed to being one through five on one side and 6 through 10 on the other side it's going to be 1 two 3 4 5 6 7 8 9 10 now with these resistors here that aren't installed that's going to keep this thing from working but if we look at the ISA bus there are tons of 4.7k resistors installed here between you know whatever the logic is on the motherboard and the expansion header so I'm just going to do the same thing on this once we figure out what the pinout is and to get 4.7k resistors I went ahead and I just uh found this junk board that's an old router or whatever and it turned out that the resistors that were right here were all the 4.7k that I needed and there they are inside here ready to be installed on this board now when we look closely at this we can tell that there are looks like four pins that are grounded these two right there and those two but luckily that leaves enough for C GA because we need RGB that's three plus intensity that's four plus horizontal and vertical sync so that's six total and that matches up with the number of resistors that we have here which is also six and yeah if we look over here remember we're only getting RGB and I guess the two sync signal anyhow I think the next step is I'm just going to install the resistors onto here and then we can scope out the pins here to try to figure out what is what we have all the 4.7k resistors installed and I did double check that none of them are shorted and yeah sorry is in the microscope and the soldering is really ugly but I just did this by hand and it's absolutely good enough so now it's time to figure out which pins are which now with the system just powered on I'm not going to be able to tell which are RGB pins but that's fine we can figure that out later the intensity bit should be able to be seen and then of course we have the two sync lines and these are the CGA signals right here pin 7even composite sync is not actually used on a regular CGA monitor I think pin 7 is actually used for monochrome video but on the commodor 1084 in in analog RGB mode pin 7 is used for composite sync but in this case we won't need to connect that at all now the pins should be active now that the resistors are in place so this is pin three and that is a snc signal 15.7 khz so horizontal sync perfect and then the next one over is going to be vertical looks like and if we zoom in a little bit more 59.8 so that is the horizontal sink oh I'm sorry I think I just misspoke pin three is actually vertical and horizontal is pin 4 pin five and six are both grounded I can see that on the PCB then we're on pin seven and then this is going to be a video signal I'm just going to have to make an assumption that this is going to be red probably just whoever laid this board out did that as red the next one looks the same so we'll say that's green and then the next one blue I guess and then the final pin 10 uh doesn't really have much on here so this is going to be the intensity bit it's probably because of that bios boot screen there's only a little bit of text that actually has intensity enabled so that actually breaks down right here and I made a Google sheet here so we have CGA on the left and book 888 on the right and there it is intensity pin 10 and I think this is going to be how we hook this up to the rgbd HDMI so I'm just going to make the assumption that is RGB like that 78 and N what I'm going to do is use this little ribbon cable here I only need seven conductors because it's RGB intensity hsync vsync and one ground line there's not a lot of load on the digital signal that's why these 4.7k resistors are just fine so this is going to be perfectly fine and I'm honestly going to just solder those on to the pin headers here and I'm just going to drill a hole in the case CU you can kind of see that it looks like there's space right there for this pin header now if you want to do this in a nice way buy the appropriate narrow pitch pin header here kind of like the ISA connector right here and use a Dremel or nibble or whatever make a perfect little square hole and then you can have this actually header EXP closed there once you've installed these resistors and then you just need to make a little adapter that goes from this to the de9 or your monitor cable or whatever but for me personally I don't really care I'm just going to drill a hole in the case right there I'm just going to run this ribbon cable right out and then I'll just attach a female de9 connector with a little hood on it and just have that hanging off the back of the computer I really don't mind that that it hangs off like that I know that's really going to bother some people but for me personally that's completely perfect and then I can just take a normal male to male 9 Pin cable plug that into a monitor or even take a CGA monitor that's built-in cable and just plug it straight into the de9 and we're good to go I don't need to have a dongle that I'm going to lose or whatever it's just going to be connected all the time all right there we go the ribbon cable has been connected I double check that none of the wires are shorting into each other I haven't yet connected anything to the other end but I'll just do that in a moment and bottom side of the board it's kind of ugly but you know what it's going to work just fine there we have it the de9 is connected now I may need to swap around the RGB lines so I'm not putting the hood or anything onto this yet but we can actually test this out with the RGB to HDMI so this cable here goes to the RGB to HDMI and is wired up for normal CGA style input and it looks like we're already configured for CGI so let's turn it on to see what happens well seemingly nothing Well RGB HMI is telling us that no sync is detected so let's poke around on the outputs while it's connected and see if those 4.7k maybe those are too high those resistors and it's pulling the signals down too low okay I'm going to go right off the N9 pin connector so that should be red oh yeah that's going to be way too low well clearly that's not going to work uh that should be green that's blue and if we flip this over and we look at the two sync signals yep we're getting 1.8 volts there on that SN signal it's way too low and this one also 1.9 volts and the intensity bit yeah just a little bit we're getting that low signal because there's a load on those lines and the 4.7k I guess is just too much now there's a version of the RGB HMI that has a buffer in it so it has like a bus transceiver on the input stage I think this version though that I'm running it just feeds these signals directly into the cpld now the cpld that's on there is a 3.3 volt part so it should be fine with that but maybe there is some type of termination resistors inside the board on the RGB HMI I can't remember how it's designed so what this tells us is those 4.7 K are just too much resistance all right it's the next day now and I've sort of molded over the problem that we were having here with those resistors causing the TTL levels to be too low my biggest problem with this particular device is it doesn't appear that the TTL outputs are buffered in any way one of the large ic's on the main board that generates the CGA Graphics seems to be what's outputting those signals and I'm not really comfortable hooking those directly up to say an external monitor using just zero ohm jumper links I was reading through that thread I showed earlier in the video a little bit more about people's experience hooking it up to CGA devices and it seems like people were just putting zero ohm jumper links there and hooking it directly up to things usually in RGB to HDMI that's probably completely fine but you just don't know with a monitor hooking it up what that's going to do and you might be putting way too much load on that IC and it could damage the outputs once those outputs are damaged all bets are off and you're not going to have any display capability out of your book 8880 so I decided to go take a look at well the originator of CGA which is the IBM CGA card at the schematics for it to look at exactly how IBM did it IBM freely published the schematics for the CGA card I'm pretty sure they did in the technical manual but luckily someone has recreated it here on GitHub which is a lot easier to read so there's a PDF you can download which actually has this nice recreated schematic and it's down here on page six we have the relevant section that I wanted to look at so the various TTL signals for RGB intensity and the horizontal vertical sync are generated by various chips on the CGA card and then they all make their way to this single LS 244 the 244 unlike the 245 which you see a lot on various old retro Hardware is instead of being a bus transceiver like the 245 which is bidirectional this is a single directional buffer essentially so inputs come into it and you can enable and disable the outputs and that's it it just passes right through and it just essentially creates a really nice stable output that on the IBM CGA card right here comes out and then just goes directly to the de9 connector they did attach some capacitors here looks like whoever created the schematics here says that their particular card had 47 picofarad this is a little bit of smoothing it's not really there for anything else it's not like a protection or anything like that so yeah IBM decided that it was totally fine just to go directly to the monitor so if it's good enough for IBM I think it's going to be good enough for the book 8088 and I want to try to replicate this by getting a little a surface amount 244 and putting that on the board and removing those 47k resist I put on here or 4.7k resistors and then just connecting the output of the 244 directly to the wire that's going to that de9 which I've already attached so I don't normally keep surface mail components in stock but I keep a bunch of junk boards and I went onto this old network card here which is dead and I pulled off a 244 which is right here and I've have test fited onto the book 8088 here and I decided I'm just going to stick this right on the board here so I just need to take the outputs of this chip for instance here's pin two which is an input and the output here is on pin three I need to conect that pin over here either to that pad right there that would have been the other side of that resistor or I could just connect it right there and I did actually managed to get some of these pins sort of bent them over connected up to some of those pads although there are still two pads that are not connected to the IC and I think that's because I'm going to have to connect them up to input pins that are on this side the way this chip is wired up some of the inputs are on this side some are on this side the outputs are on the opposite side so it's a little bit of crisscross thing going on but it's nothing that I couldn't do with a few bodge wires and like I said hooks up to power there's probably going to be 5vt power right over here on this Isa bus connector add a bypass resistor for the power input and then we should be good to go with a nice solid 5volt TTL output right on here while not affecting anything to do with the signal that's going to the LCD screen okay there it is it's all installed and it's really ugly but I think it should do the trick so 5 volts I'm taking right off here off this pin and that goes over here to the 5vt input here on pin 20 pin one of this IC is up here at the top this is the ground pin and I did put a bypass capacitor right here and I'm picking up ground right there from the header so it goes there to that pin and then we have to ground some additional pins on this chip so I'm not using two of the inputs those are grounded and this is pin 19 which is one of the enable pins that is grounded in addition in pin one here should also be grounded so that's also grounded they're all sort of common right here on this capacitor I decided not to use any of those pads on the left side where those resistors were I just went straight to the connector there and I mean it's really hard to see but I just sort of layered this all up on there and I think it's all good I went ahead and I double check that nothing is shorted out and doing a little bit of threedimensional work here you can see the ground lines there are lifted up and meanwhile the ones that have signals are on the lower level now I'm going to be using some hot glue to just sort of attach this whole thing down since it is just sort of flopping around but I think in the meantime we're ready for PowerUp testing just to see if anything smokes hopefully no damage occurs when I power this on okay so the oscilloscope is powered up it's connected I'm ready to go I just want to start seeing if we're seeing good 5vt TTL levels out of the de9 connector let's hope that I turn this thing on it does Power Up Normally here we go good sign everyone good sign all right so we're looking at the blue pin here on the de9 and it looks good ah this is the green pin that doesn't look so good ah and that's the red pit and that doesn't look good either okay let's power this off I need to try to figure out what exactly is going wrong here all right I think it's fixed I found a couple issues I had wired up two of the pins wrong on the output side the I think red and intensity or or two of the pins yeah were swapped but there was also a short from one of the outputs to the inputs which is kind of kind of bad because I was putting 5 volts back into the logic chip on the board so hopefully it was able to withstand that so let's power it on now and I'll take a look at the output on here and let's see if we see 5 volts and the correct signals on all the pins all right we're on the top row here so this is going to be blue then this pin here is green that looks good and red that looks good as well and if I flip this over this is the intensity bit okay looks good there's a little bit of noise down at the bottom but that's completely fine this is TTL vertical sink and horizontal sink on the last pin nine okay this looks good now so all it's left to do now is hook this up to the rgbd HDMI we should have a working signal and if it does work I'm going to grab an actual CRT Monitor and then we'll see if that works cuz that's the ultimate test all right the RGB HMI is connected up see what happens okay maybe I have the sink lines reversed that would explain what we're seeing here we're seeing the correct colors through all this mess here ah yes okay taking a look here I remember pin 9 was horizontal sync and pin 8 was vertical sync so I just need to swap those around and I'll do it on the de9 connector that's the easiest place to do it all right doesn't look like much but I swap those two lines around you can't really tell cuz it looks exactly the same there we got it it freaking Works awesome all right so fully buffered CGA output and look we can see that static there and I think we have yep so the intensity bit is absolutely working because you see at the top there the a F6 a f8 etc that's fine and then the colors there look perfect as well the book is nearly back together I don't have all the screws back in I just have some of them as you can see I drilled a little bit of a hole there so the cable can pass through I'll just dab a bit of hot glue into that so it doesn't get tugged on if we open this thing up it all looks good here I haven't actually turned this on yet but um hopefully everything is working I did put the sound card back in as well all right let's power this up okay good it's powering up okay good the screen is still working I was uh a bit worried there for a second we had a little bit of a delay and there we go we have CGA output as well um I don't know why you can't hit control delete oh you can I thought I did that earlier and it did not work well yeah looks good there we go I think the colors are good as well so here we are in Jim Leonard CGA compatibility suite and as you can see it is Rock Solid the video display let's check out color here just to make sure everything looks good yes it looks really good so the color dark yellow is actually Brown on CGA last time I mentioned that people asked about it so you see the top there it goes blue green cyan red purple these are all dark colors and then it has what is dark yellow but notice it's rendering as brown BR and it works that way on CRT as well there's a bit of extra logic in the color decoding circuit in the monitor that basically adds a little bit of extra red to the dark yellow to make it that brown color because otherwise you'd have the light yellow down at the bottom and then that brown which is right about there would just sort of be the same shade of yellow but just darker and it looks like on the LCD they're attempting to add a little bit of extra red into the dark yellow there to make it a bit more Brown it doesn't like quite look correct it's sort of a orangey color if anything not that Brown is not just dark orange but it just doesn't look correct what you're seeing on the video capture is actually more the correct color next I just want to quickly test that this adlib card here is working because that's something we actually haven't tested so wow look how slow that is we will do three for adlib yeah it totally works and it's coming out of the internal speakers here on the uh computer and sure enough function F1 does mute the audio and turns it back on again so that's pretty cool so the one thing that I can confirm now is that the beep speaker is on the motherboard and the opl audio is separate it would have been nice if those were combined together into these speakers that way for games that use both you would have that combined sound my assumption is that the audio output Jack on the side of this thing also only outputs the opl sound and not the beep speaker a little bit unfortunate all right before I do any further testing on this thing I'm just going to put it together the rest of the way give it a little bit of clean on the outside put a little plastic cups back on and then we can actually run this through a few other other tests all right I have the Commodore 1084 sitting on the bench this thing has TTL input which it's currently set to the book is all back together and I have it of course connected to the Monitor and this thing is fully reassembled I did go ahead and remove the stickers from this thing I'm not a fan I am not a fan of stickers anywhere on any laptop including old retro ones like this so yeah I peeled those off all the little plastic doors are back on front and bottom or top and bottom I also put little rubber feet on the bottom of this thing because it didn't have any so now when it sits on the bench it's well it's nice and grippy so will this work with a real 1084 monitor let's find out here we go oh dear well that's interesting that is interesting indeed so we have text here but we are getting basically a green background and it was working perfectly on the RGB to HDMI as we saw in the video capture so let me go hook it up to that see if that is still the case all right well I obviously did something and broke it because the green isn't working at all on the RGB HDMI sorry it's not focusing on it it seems that I caused this problem both when I reassembled the case and stuck some hot glue into this and also into the back of the computer and I obviously dislodge something let me quickly try and fix that and absolutely when I took the computer apart which it's pretty easy now that I've done it a few times I noticed the green wire had just come disconnected and that was because I was sort of pushing the ribbon cable through the whole I drilled the case and it was pushing up against those connections and it just pushed the green one I guess I didn't have a very good solder joint so I just reconnected it put it back together let's see what happens now come on oh yes there we go it's working awesome so the fact that I have an LS 244 bus driver inside the machine for the CGA output means that a it should protect if anything bad goes on with that connector that's hanging out the back of the computer means it hopefully she'll protect that logic IC that's sitting on the motherboard that's generating the CGA signals it's almost like that 244 it's a sacrificial chip okay so we're booted up into Landmark speed test this is version two and if I hit function F6 we go from 2.27 to 3.81 MHz so that's that turbo speed in effect now if we type di in here the scrolling speed is still incredibly slow and if I push function six while it's in the middle of scrolling I honestly don't even notice any difference whatsoever so whatever routines are happening in the Bios that prevent the snow or the flash anger in the dir make this scrolling speed unbelievably well brutally slow to be honest now I'm going to run the CGA game called Magid duck 1.0 and this is pretty cool and that it uses the special CGA mode where it reprograms the rose and you get a pseudo 16 color Graphics mode press y to start the game now there may be a ton of snow maybe not actually to be honest oh you know there won't be any snow because this is running in 40 columns mode and the snow only happens in 80 columns mode so if we go to new game we do slot one there it is check that out no snow really fast animation even on this slow old computer so this is like a little platformer now it's equivalent of using pety text mode essentially to uh well give us this 16 color mode even on Old CGA XTS and it runs really fast which is frankly pretty awesome and um let's see I can fire you're supposed to collect the gems and things like that and well yeah it's really good now doesn't this just look so much better hooked up to a monitor than this garbage it's really let down by this terrible screen which is really a shame because I've played with plenty of those like little handheld Nintendo consoles and things that have far better screens than this now I'll be at their much smaller than this but you think a 4x3 screen that would have looked a lot better than this would have been available for this little machine and considering plenty of work went into designing this I mean not just the PCB but the plastic case and well the physical design of it all it's quite good actually I mean the build quality is not terrible for this little cheap computer now magad duck was not the only game that existed out there that used that CGA mode and this is another one oh but it uses 80 columns mode so we got got the snow and it's paku paku which is an awesome Pac-Man clone for IBM XT and 5150 machines and unfortunately while the snow oh it even has decent sound oh it's using the adlib okay well I didn't even know hey the slow the snow is very minimized though all right well it has beep speaker sound as well if you don't have an adlib but check out how great this is I mean it looks absolutely f fantastic to my eyes and even ignoring that little bit of snow at the top which is honestly very minimized now the reason why I know it's using adlib sound is because the sound is coming out of the speakers around the bezel of the screen all right I got them all now we saw that post about a hardware mod for this thing to restore some kind of Cycles or something like that I think what it's talking about is that this doesn't do the dma based Dam refresh like the original PC therefore this runs a little bit faster and it's not cycle EXA accurate so with that mod done I think it makes this thing 100% cycle accurate versus the very first 4.77 MHz machine and then it's quite possible this game is so perfectly tuned for those original machines that the snow is completely minimized and is held outside of the visible refresh area of the screen and I guess the timing is maybe slightly off on this which is why most of the time like right now there's no snow and we just get a little bit of a band that runs through the picture all right that is going to be it for this video I played around with this thing enough to know that well I think it's a very interesting idea but there are so many limitations of this particular version of the book 888 I really can't recommend it now I've heard that there's a version two that came out that has some additional features like maybe even EG support built into it I haven't looked at that so I can't really speak to it but this first version it would have been so nice if they had just included the CGA output on the system using the buffer and everything that would been so simple to add that in like with just pennies of additional cost I don't know why it wasn't done because if it was it would have been so convenient for using an external screen like this as far as repairing this machine goes I really think I had a lucky break with the fact that it worked some of the time especially when I took it out of storage after a few months because that enabled me to kind of poke around the motherboard and figure out which chip was bad I'm not 100% sure that I could have figured that out if it wasn't for that although I suppose it would have been just a good idea to go ahead and swap at least all the socketed chips at least the ones I had spares for and in the case of this machine it would have made it work again if you have one of these machines that didn't work in the same way and you're getting a good five volts when you turn it on to those logic ic's you may as well go ahead and try to procure some of those chips at least the socketed ones and swap them out one by one to see if any of them make the computer work but in the case of the 8288 which was definitely a problem on this board it probably worked when this thing was fully assembled and tested at the factory or whatever but then once it got shipped out the little Bond failure that was happening inside that IC just got worse and worse now as I mentioned earlier that 8288 chip was pulled off another motherboard like a scrap motherboard that was probably ewed a long time ago and that means that those are removed from boards in who knows what way like with blowtorches or whatever and that can damage the bonds inside the chips and can lead to some of these types of failures with all that said thanks very much Ezra for sending in this computer really glad I've been able to work on it and modify it to make it better by adding this external CGA would be kind of cool to add a toggle switch to the side of this thing that would shut down the internal screen which would probably make the battery last far longer I think most of this stuff doesn't take up a lot of power and the screen is probably the biggest consumer of power inside this machine and obviously if you're using it with an external monitor well you don't need that internal screen anymore anyhow if you like this video and found it helpful a thumbs up would be much appreciated but if you didn't you know what to do huge thanks to my patrons their names are scrolling beside the screen they get early access to videos and occasional live streams and things like that so you can check out the link in the description below below if you want to become a patron my patrons make it possible that this is my full-time job and without them really wouldn't be possible so I'm very very grateful to them and I really do appreciate the support hit subscribe if you haven't already and check out the second Channel and all that other good stuff and I guess that's going to be that so stay healthy stay safe and I'll see you next time [Music] bye-bye

Channel: Adrian's Digital Basement

Views: 78,073

Keywords:

Id: DAPCi-TQiy8

Length: 70min 14sec (4214 seconds)

Published: Sat Apr 13 2024