Jump to content
IGNORED

7800 Restoration


juansolo

Recommended Posts

17 minutes ago, DrVenkman said:

The RF on my NTSC 7800’s is terrible. Just rotten. The improvement from composite through the UAV is amazing, and the S-video is even better. Not sure how much difference it will make on your test systems but I’m looking forward to seeing your appraisal. 

 

The RF on my NTSC A1 is so good that I don't feel any pressing need to mod.  It's plugged into a DVD recorder, which then passes the video to my 60" UHD television via component video. Perhaps there is some signal processing going on in that mix, not sure.

Link to comment
Share on other sites

48 minutes ago, DrVenkman said:

The RF on my NTSC 7800’s is terrible. Just rotten. The improvement from composite through the UAV is amazing, and the S-video is even better. Not sure how much difference it will make on your test systems but I’m looking forward to seeing your appraisal. 

That was part of the reason I wanted to do this in the first place is to see what our mod does vs the UAV, vs the UAV plus chroma switching. Our mod is fairly basic but it's a lot better than the simple composite mods out there. The UAV does some really nice things and generally speaking does a better job at most things, but it's still blending the chroma so it'll be interesting if the switching makes it even better. We can't compare to RF because we don't have anything with an NTSC tuner in it sadly. But generally speaking, from the many different consoles we've modded, RF rarely looks good next to a decent composite mod.

 

I would say though that the NTSC 7800 is vastly better out of the box than the PAL one in terms of picture quality.

Edited by juansolo
  • Like 1
Link to comment
Share on other sites

I make no appologies for the tarty piccs that are about to occur. All can be embiggend with much clickage.

 

Amerkin-17.thumb.jpg.2efa80fdba05a5b981758c78b67f9536.jpg

 

It's complete, for now, and I'm really happy with the results. The picture on it is better than my PAL machine that has a UAV with chroma switching, and there's just our AJM in the NTSC machine. Looking forward to seeing what changes the UAV then UAV plus switching might make.

 

Amerkin-19.thumb.jpg.7856b7f8fbe87a40c3da2426a180adcc.jpg

 

It's not the neatest install, but as it's only temporary and this things is likely to go through several iterations before I'm done with it, it'll do for now.

 

Amerkin-18.thumb.jpg.8714a99b8f0c7b70c6de6cfdfde40fe9.jpg

 

I wouldn't normally do the outputs like this. Normally there'd be a case mounted 3.5mm stereo jack carrying both composite and audio, with the S-Video also just bolted into the case. As I'm planning to be taking this apart a bit, I thought I'd make that as easy as possible. There's no composite out and there is a board mounted phono carrying audio where the channel switch used to be. You need to drill some holes in your PCB to do this and cut two traces. Also I've grounded and anchored the S-Video jack using twisted solid core wire to the RF modulator mounting holes, this ended up being surprisingly rigid and the socket doesn't move. It's still bolted through the case when it's in, but that's super easy as it all just lines up when it's dropped in there. It's a bit of a faff to do, but in the long run it'll save me loads of time.

 

Amerkin-21.thumb.jpg.882c81863df9d22ca565058de99c92d4.jpg

the new socketry

Amerkin-22.thumb.jpg.a67f7c9af6dce5fc0a84b7c6d483a9f2.jpg

 

So that'll be it for a while, when the UAV lands the experiment will contine. But for now, we're done.

 

Amerkin-20.thumb.jpg.f2ee9539778715fee4b85367c657d100.jpg

Edited by juansolo
  • Like 4
Link to comment
Share on other sites

Hit my first bit of oddness with the NTSC machine, seems it has the sticky fire button problem noticable on Rikki and Vikki resulting in your player auto picking up boxes in single player with only one gamepad plugged in.

Now this can be gotten around by plugging in a second controller. Which is fine, but I only have two controllers (a gamepad and a modified 7800 Zipstik) for both my machines so I'll probably end up making a dongle for the 2nd port. This does however potentialy scupper my internal Atarivox plans. I'll have to find out what lines the AVX actually uses and hope they're not the ones we need to add resistors to to negate this problem (if anyone knows this'd be VERY helpful as my two are buried inside machines). If they are, I can potentially negate the issue when running the AVX when it's switched in, and it'll just automatically negate it when it's switched out and I'm running two player with a second controller. It's a potential work around that depends on the planets aligning a little. That said the AVX itself might make the problem go away. I guess I'll be taking my 2600 apart and extracting that to test that.

 

First, of course, we're gonna swap some parts and see if we can't fix the problem. A job for Wednesday when I'm next over. Now you can see why I made it super easy to dismantle ;)

 

As for the colour fringing issue, this is interesting. Looks like a UAV equipped NTSC 7800 does this also (check out Dungeon Stalker):

 

 

Dungeon Stalker and Frenzy were the most apparent ones to us, but you can see it on certain colours in the palette picture. That'll be interesting because that looks awfully similar to my AJM equipped machine!

Edited by juansolo
  • Like 2
Link to comment
Share on other sites

7 hours ago, juansolo said:

Hit my first bit of oddness with the NTSC machine, seems it has the sticky fire button problem noticable on Rikki and Vikki resulting in your player auto picking up boxes in single player with only one gamepad plugged in.

Now this can be gotten around by plugging in a second controller. Which is fine, but I only have two controllers (a gamepad and a modified 7800 Zipstik) for both my machines so I'll probably end up making a dongle for the 2nd port. This does however potentialy scupper my internal Atarivox plans. I'll have to find out what lines the AVX actually uses and hope they're not the ones we need to add resistors to to negate this problem (if anyone knows this'd be VERY helpful as my two are buried inside machines). If they are, I can potentially negate the issue when running the AVX when it's switched in, and it'll just automatically negate it when it's switched out and I'm running two player with a second controller. It's a potential work around that depends on the planets aligning a little. That said the AVX itself might make the problem go away. I guess I'll be taking my 2600 apart and extracting that to test that.

 

First, of course, we're gonna swap some parts and see if we can't fix the problem. A job for Wednesday when I'm next over. Now you can see why I made it super easy to dismantle ;)

 

As for the colour fringing issue, this is interesting. Looks like a UAV equipped NTSC 7800 does this also (check out Dungeon Stalker):

 

 

Dungeon Stalker and Frenzy were the most apparent ones to us, but you can see it on certain colours in the palette picture. That'll be interesting because that looks awfully similar to my AJM equipped machine!

Yes the color fringing is something I've seen on all 7800s I've put UAVs into. That is a lot of 7800s at this point. However, it doesn't seem to happen or show up as prominent with 2600 games so I suspect it is more of a MARIA issue vs the UAV? The issue with the stuck trigger is the TIA chip. I chose to live with mine doing it and haven't replaced the TIA to try and correct for this. I first noticed it when I was having a really good session of Xenophobe going and suddenly, without warning, the second player started up on the lower portion of the screen? I asked about it in these forums at the time and was advised the TIA was the issue. 

 

Anyway, I state this because I've plugged in and been able to use with my Avox without any issue. But, I did have to remove the PIC off it (Disabling Vectrex use) in order to do so as prior to that, it would reset itself and stopped working within a few minutes otherwise.

 

  • Like 1
Link to comment
Share on other sites

I've got a later AVX so I shouldn't need to de-PIC it thankfully. But if it addresses the problem, it kills two birds with one stone for me. For the time being I'll make a ghost controller dongle for the P2 socket until I can get another AVX and fit it internally. Then when I'm single player and it's on, things should be ok, and when two player, I should also be ok. In the meantime if I end up with spare TIA at some point, I'll swap it out.

 

I didn't really see the fringing at all on the 2600 side. It's an exellent picture all-round other than that also. Really surprised by just how well our relatively simple mod worked on it!

  • Like 1
Link to comment
Share on other sites

Assorted Wedge Hardware Tipz...

 

On 6/7/2021 at 12:24 PM, -^CrossBow^- said:

Made in late '87 and essentially uses the newer revision board with the extra timing circuit silkscreened but bypassed completely. 120ns Sony RAM with the standard resistor off the WE line as well. I think I've only seen that resistor installed on the 120 and 100ns variants.

This resistor messes with /WE's slew rate so it'll only be there for faster grade memory - to make it react slower.

 

 

On 6/11/2021 at 11:01 AM, juansolo said:

We can't compare to RF because we don't have anything with an NTSC tuner in it sadly. But generally speaking, from the many different consoles we've modded, RF rarely looks good next to a decent composite mod.

 

I would say though that the NTSC 7800 is vastly better out of the box than the PAL one in terms of picture quality.

From experience, the NTSC boards have marginally better signal integrity (including the digital junk) but both are kinda noisy and the layout isn't optimal. So the picture quality over RF is gonna be all over the place.

 

 

7 hours ago, juansolo said:

Now this can be gotten around by plugging in a second controller. Which is fine, but I only have two controllers (a gamepad and a modified 7800 Zipstik) for both my machines so I'll probably end up making a dongle for the 2nd port. This does however potentialy scupper my internal Atarivox plans. I'll have to find out what lines the AVX actually uses and hope they're not the ones we need to add resistors to to negate this problem (if anyone knows this'd be VERY helpful as my two are buried inside machines).

Going by this documentation, the AtariVox doesn't touch the Paddle / ProLine pins (5 + 9), so you should be able to add pulldowns to these without messing up the AtariVox's functionality.

 

What you might want to consider implementing is a circuit where these pulldowns are only enabled while in two-button mode. So gated through a transistor and using TIAEN and RIOT PB2 / PB4 to enable it - similar to the Q5, Q6, Q8 pullup network. This would (or moreso should) avoid interfering with the charge and discharge timing when using paddle controllers.

  • Like 3
Link to comment
Share on other sites

33 minutes ago, TailChao said:

Going by this documentation, the AtariVox doesn't touch the Paddle / ProLine pins (5 + 9), so you should be able to add pulldowns to these without messing up the AtariVox's functionality.

 

What you might want to consider implementing is a circuit where these pulldowns are only enabled while in two-button mode. So gated through a transistor and using TIAEN and RIOT PB2 / PB4 to enable it - similar to the Q5, Q6, Q8 pullup network. This would (or moreso should) avoid interfering with the charge and discharge timing when using paddle controllers.

Very useful stuff, cheers!

Link to comment
Share on other sites

Thought we'd do a bit of due dilligence when I was over today. So with R&V the problem manifests when the pad is plugged into the left hand port. Plugged into the right hand port on it's own, it goes away. Likewise if you plug two controllers in, both work fine. So that's exactly the symptoms we expected from other's who've had the problem. Just to be absolutely sure we swapped out the associated 3 transistors and checked the associated resistors. All were fine.

 

So a TIA issue, I don't have an NTSC TIA so it's not something I can just swap out, so I'll be bodging the fix for this. What I'm going to do is put in the switching for the internal AtariVox (which will be happening on our next order when Al does the next cycle of new stuff). When I'm running the AVX I'll be running single player so it'll also switch in the resistors to make the left port play ball. When I'm playing two player, I have to switch the AVX off anyhow, but it'll also switch out the resistors. Nice and easy fix that is only two resistors more than what was already going into the machine regardless.

Link to comment
Share on other sites

Well the UAV arrived and somehow managed to avoid any additional charges, so my fears of a super expensive UAV, ended up making it the cheapest one I've bought in a long time at £30 or whatever it was. Nice.

 

Amerkin-25.thumb.JPG.b2c5e168379af99ad373748359372210.JPG

 

I had a spare terminal block from one of the many other UAVs that've ended up in things so dropped that on for the outputs. The other wires I solder kinda differently... There's one thing I really don't like about the UAV, I understand why it was done, it really is a marvel just how small it is, but the solder pads when you end up putting it into a 7800 are just too small and there's not enough pad around them. To the point that here I decided to solder to it as I would to a via. Which worked really nice and means that should I need to remove it, and re-attach it elsewhere it would be easy.

 

Amerkin-26.thumb.JPG.483731d624d620ed7d50d3d42916f62f.JPG

I added some hot snot under the S-Video jack just to make it super rigid when testing out of the box

 

There is obviously something happening on the seperate chroma inputs that's specifically for the NTSC 7800. We tried to trace them to a point but couldn't work out where it went after hitting the IC on two seperate pins. I kinda lost interest and decided to put it in at that point and see. What's interesting is that the colour fringing, as seen on our AJM mod and the ZPH videos remains, but the picture is noticably sharper, which is what we expected. Thing is I don't think the NTSC machine has the issues with chroma blending that the PAL machine has, to the point that I'm fairly convinced that adding switching would make next to no difference. I may yet try it, for science, but I think that quite frankly until someone comes along with an RGB mod, if you've got an NTSC 7800, the UAV just does the job better than anything else out there currently.

 

Time for some test cards:

 

Amerkin-36.thumb.JPG.2cb20af58545d3ac5218dd2dff7bb699.JPG

UAV into 14" PVM via S-Video (7800)

 

Amerkin-35.thumb.JPG.467d680261044c4f54d98ff334c7740f.JPG

UAV into 14" LCD via S-Video (2600)

 

Amerkin-34.thumb.JPG.061d790104b914691e2f104c3de3e00f.JPG

Ballblazer into LCD

Edited by juansolo
  • Like 3
Link to comment
Share on other sites

Some panel vs CRT comparison shots. Usual caveats apply when it comes to photographing such things:

 

Amerkin-27.thumb.JPG.d8d808f0e62938de8b3215246f857516.JPG

CRT above - LCD below

Amerkin-28.thumb.JPG.83cd33f930d1c1dd49ce0a7983fab09e.JPG

 

Amerkin-29.thumb.JPG.85d3a982218959b39cce79b86d20881c.JPG

CRT above - LCD below

Amerkin-31.thumb.JPG.85948c1a7b56c1f6a872e3a120178007.JPG

 

Amerkin-32.thumb.JPG.8055f97766673d48aec574f09e39477c.JPG

CRT above - LCD below

Amerkin-33.thumb.JPG.8acdb26ff89ea1b337371aa963e4805f.JPG

 

FWIW, it looks really rather lovely on the CRT. The fringing appears to be here to stay on both so not a lot we can do there. The panel is fine, but wouldn't be my choice to plug this into.

  • Like 3
Link to comment
Share on other sites

You know...the more I look at the color (colour) fringing issue. I wonder if it has to do with the timings on the 7800? The 2600 RGB mod will produce similar fringing in some games in some parts of the screen due to changes it makes in the timing of the video. Only in the case of the UAV (and other AV mods I've tried), it is coming from the maria side since the issue is less seen on the 2600 side. Could it just be a flaw in the maria design causing this?

 

Additionally you have confirmed that the issue isn't due to the RF modulator components either because you have all of that removed and yet it still remains. So I'm not sure how much cleaner a signal you can input in to the UAV beyond this.

 

  • Like 1
Link to comment
Share on other sites

10 minutes ago, -^CrossBow^- said:

Additionally you have confirmed that the issue isn't due to the RF modulator components either because you have all of that removed and yet it still remains. So I'm not sure how much cleaner a signal you can input in to the UAV beyond this.

Totally aggree. We were debating the worth of the chroma switching given the results... Might do it at some point, but I seriously doubt there'd be any noticable benefits to it.

 

10 minutes ago, -^CrossBow^- said:

You know...the more I look at the color (colour) fringing issue. I wonder if it has to do with the timings on the 7800? The 2600 RGB mod will produce similar fringing in some games in some parts of the screen due to changes it makes in the timing of the video. Only in the case of the UAV (and other AV mods I've tried), it is coming from the maria side since the issue is less seen on the 2600 side. Could it just be a flaw in the maria design causing this?

Likely given we noticed that the colour fringing appears to be a static version of the aberations we get on PAL machines with the colour rolling, which we tracked to the MARIA because they're temperature sensitive!

Edited by juansolo
Link to comment
Share on other sites

5 minutes ago, juansolo said:

Totally aggree. We were debating the worth of the chroma switching given the results... Might do it at some point, but I seriously doubt there'd be any noticable benefits to it.

 

Likely given we noticed that the colour fringing appears to be a static version of the aberations we get on PAL machines with the colour rolling, which we tracked to the MARIA because they're temperature sensitive!

Not just the Maria. The TIA and GTIA chips also change their timings slightly depending on temperature and this can be seen in changes to the hues. @Trebor is the resident expert in this field.

But yeah, the right hand side color fringing on some of the colors looks nearly the same as what happens with timing issues on 2600s with the RGB mod in place. Also, using your CRT (Although you can't get it quite perfect), you can adjust that trimmer on the UAV to adjust the overall phasing and artifacting. I will sometimes mess with this using Tower Toppler on a CRT through composite output (Since s-video removes the artifacting completely). But the colors still won't be quite right compared to RF output. Not sure if anything on the UAV can be changed (Resistor values and the like) to affect it but just something else to play with since you have it all in place.

 

  • Like 2
Link to comment
Share on other sites

On 6/13/2021 at 3:10 PM, -^CrossBow^- said:

The issue with the stuck trigger is the TIA chip. I chose to live with mine doing it and haven't replaced the TIA to try and correct for this. I first noticed it when I was having a really good session of Xenophobe going and suddenly, without warning, the second player started up on the lower portion of the screen? I asked about it in these forums at the time and was advised the TIA was the issue.

Out of intrest does Tower Topler constantly return to menu when you try and start a game? I'm assume it's the same issue. Again no big problem as the game is just pure evil.

Link to comment
Share on other sites

3 hours ago, Trebor said:

The right side Difficulty Switch controls that feature.  Switch it to the right, instead of to the left.

 

Not so evil:

 

 

;)

Ah, that explains a lot as adding the second controller in that scenario wasn't changing things. Back to my original plan of adding some resisitors on the Atarivox switch that's happening shortly.

  • Like 1
Link to comment
Share on other sites

Finally got around to adding the chroma switching in and it had the expected results. In that if it's any better (which logically it probably is) it's by a small amount and it certainly not worth the effort. What I do now have is a switched chroma signal so it's as clean as it possibly can be. Which is handy to know for experimenting.

 

Amerkin-37.thumb.JPG.df64112895477a0a3bc5af2dd59b36d2.JPG

 

There was also a bit of future prep in the form of socketing the BIOS. The biggest PITA was that the IC was jammed so hard into the PCB it was super difficult to get out without lifting a couple of pads. So had to dig out the kynar to fix those. Same on the other one, though just one that time, you can see the fix for that one. Thankfully they're under the sockets so won't cause any further problems.

 

Finally a bit of experimentation in the form of socketing the 74AHCT32, which is an OR gate for the luma signals. @marauder666 has a theory on the colour fringing he thinks it might be down the chroma or the luma timings are slightly out from one another. Interestingly it's not as bad on the 2600 side. So the AHCT has a fast 5ns propogation time (interestingly the PAL machine is a LS32, which are about 10ns). We've tried removing the IC and hard wiring it (essentially making it 0ns) and that made no noticable difference, so we'll try to slow it down and see if that does anything at all. Just having a play to see if his theory is up to anything.

Edited by juansolo
Link to comment
Share on other sites

This update was brought to you by me being a bit cheeky, and @Albert sneaking an Atarivox into my missing audio CD package. Cheers dude! Pics aren't so great because I had the camera in the wrong mode, sorry about that.

 

I'm a bit of a sucker for fitting AVXs inside my Ataris, so that's what I've done here. What it's also allowed me to do is add the bodge fix for the TIA bug my machine suffers from. So when the switch is set to AVX, it also adds in the two resistors the second pad would have, meaning with only one stick plugged in, the games that are afflicted by this bug work properly. When I'm playing 2 player and turn the AVX off, the other pad would be plugged in and the bug doesn't happen. All round, a sweet little fix.

 

 

Amerkin-38.thumb.JPG.2ee7c7456cf03e1307e71fe9f5f08b39.JPG

ribbon from the back of port 2 (not as pretty as Voultar's exquisite mounds)

 

Amerkin-39.thumb.JPG.a53318ee15e246a6cad4e00b26345b71.JPG

Cable to switch with bodge applied

 

Amerkin-40.thumb.JPG.f7acad23b7afd9c605e4cdca62a2a773.JPG

AVX fitted. It's just on Scotch foam tape at the moment. Not sure how warm the TIA gets...

 

Amerkin-41.thumb.JPG.76869b2aad6af741d20edebf646d81ad.JPG

Switch action... I really hate cutting square holes.

  • Like 1
Link to comment
Share on other sites

TIA and RIOT both get pretty warm but not usually as hot as the 6502 and Maria can get. I've really thought about putting a heatsink on at the least the CPU as it seems to be the warmest component in the system next to the 7805. But that is easily remedied in regards to the 7805.

 

Link to comment
Share on other sites

7 minutes ago, -^CrossBow^- said:

You should create a simple mixing circuit and just wire the AVOX+ output to your audio out rca on the back?

Yup, we considered that when we first started doing this. The reason I didn't on the first (a PAL 6-Switch 2600) was I happened to have a really nice speaker that fit perfectly behind one of the grills. I then just kinda got used to the 2600 itself mocking me for dying, so have kinda stuck with it since. Also we had a few of these tiny speakers which actually sound good and fit easily in 7800s.

 

It would be really easy to do however. I'm sure I could knock up a vero layout for one pretty easily if someone wanted to try it.

Edited by juansolo
  • Like 1
Link to comment
Share on other sites

I'm a little concerned about heat with the speaker on the TIA & RIOT, so I'm going to blend the AtariVox into the audio out. Normally I wouldn't put the speaker there, but this machine is for experimentation so it all needs to be easily removable (it's mounted to the case on my PAL 7800), and the easiest way to do that is to lose the speaker. It all comes down to how the LM386 is being used (it's an amplifier, so not line level). If the pot is a gain control the signal will likely be too hot to blend and would require attenuation. If the pot is on the output and is already doing that, it might be possible to simply plumb it straight into the back of the output cap along with the two other outputs as essentially it's all doing the same thing and the pot will act as the resistor (it being variable will allow us to balance it with the others). It should make life ludicrously simple if it's the latter. We shall see when I take it back over to Cleggy's on Weds for a quick fettle.

Edited by juansolo
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...