Jump to content
IGNORED

1088XLD - Custom A8 Computer in 1050 Drive Case


mytek

Recommended Posts

8 hours ago, MacRorie said:

Can y'all give me the original name for this part so I can find it on the BOM?

 

IS this what we are discussing?

 

1          Various Headers         0.1" SIL Single Row Breakaway Male Header Digi-Key           S1011EC-40-ND          (Note 5) J3,J6, J7, J8,J14,J16

 

Note 5 has this to say:

Note *5 Substitute Digikey # S7005-ND for J3 on mainboard for use with Dropcheck Rear Level 1 I/O board only

 

So are we changing the part number?

I believe I will be breaking out the J3 connector and placing it in a separate line item with the choice of either Digikey #S7040/S7005 -ND  or the original Digikey #S1011EC-40-ND depending on whether the customer will ever want the Mod boards.  Mainly it's elevation from a note to an actual line item choice much like the type of socket option already listed in the BOM.

  • Like 1
Link to comment
Share on other sites

  • 3 weeks later...
Just now, AtariGeezer said:

Except for the dreamblaster s2,  the rest of my parts orders will arrive this week.

So hopefully I can start my 1st build this weekend :)

 

@MacRorie,  How many of these boards are out there???

22 sets for XLD

 

I can also let everyone here know that there is a Black Friday/Cyber Monday special on these coming!

Link to comment
Share on other sites

  • 3 weeks later...

Possible ANTIC Issue

 

I discovered this when testing with my V1.0 1088XLD board, that in NTSC mode there seems to be a PMG problem with certain Antic chips. I only have a test case of one, so I can't really say that it's a wide spread problem, but it is something to keep in mind if you start seeing PMG problems in for example the U1MB menus. I also haven't verified if the bad Antic works properly in either a stock system or an XEL.

 

This one tested fine.

NTSC_Antic_Good.thumb.JPG.f004def1447869d0f0601e6f5b84fd3e.JPG

 

This one exhibited PMG problems.

NTSC_Antic_PMG_issues.thumb.JPG.197805d1388ce7e18fceac3308937a56.JPG

 

  • Like 3
Link to comment
Share on other sites

33 minutes ago, mytek said:

Possible ANTIC Issue

 

I discovered this when testing with my V1.0 1088XLD board, that in NTSC mode there seems to be a PMG problem with certain Antic chips. I only have a test case of one, so I can't really say that it's a wide spread problem, but it is something to keep in mind if you start seeing PMG problems in for example the U1MB menus. I also haven't verified if the bad Antic works properly in either a stock system or an XEL.

 

This one tested fine.

NTSC_Antic_Good.thumb.JPG.f004def1447869d0f0601e6f5b84fd3e.JPG

 

This one exhibited PMG problems.

NTSC_Antic_PMG_issues.thumb.JPG.197805d1388ce7e18fceac3308937a56.JPG

 

The C012296 is the older version of the  Antic. I remembered it being mentioned in the RamboXL installation:

http://www.atarimania.com/documents/rambo_manual.pdf

 

Don't know the difference, but pin 8 needed to be connected to the board. Does that offer any help?

  • Thanks 1
Link to comment
Share on other sites

14 minutes ago, gozar said:

The C012296 is the older version of the  Antic. I remembered it being mentioned in the RamboXL installation:

http://www.atarimania.com/documents/rambo_manual.pdf

Yes it's what I purchased from BEST, since I was looking for lower cost and the refresh timing didn't factor in with SRAM being used. And this now rings a bell as to where I got the CO21697 version from. It came out of one of my XEGS systems, and if I recall correctly it worked fine in that system. Since the older CO12296 Antic chip is usually cheaper to purchase, it's nice that it is the one that works properly in the XLD.

 

Timing characteristics of other areas of the chips might have inadvertently varied due to the different DRAM refresh rates that were designed in between the two part numbers, this could be a contributing factor as to why I see PMG issues with the CO21697.

 

14 minutes ago, gozar said:

Don't know the difference, but pin 8 needed to be connected to the board. Does that offer any help?

Hmm... not really, but thanks for the suggestion. In the 1088XLD, pin 8 (REF) is connected to anything else that required it (i.e., MMU, ECI). But with the RAM being Static instead of Dynamic, it plays very little into this design.

 

Link to comment
Share on other sites

2 hours ago, mytek said:

Possible ANTIC Issue

 

I discovered this when testing with my V1.0 1088XLD board, that in NTSC mode there seems to be a PMG problem with certain Antic chips. I only have a test case of one, so I can't really say that it's a wide spread problem, but it is something to keep in mind if you start seeing PMG problems in for example the U1MB menus. I also haven't verified if the bad Antic works properly in either a stock system or an XEL.

 

This one tested fine.

NTSC_Antic_Good.thumb.JPG.f004def1447869d0f0601e6f5b84fd3e.JPG

 

This one exhibited PMG problems.

NTSC_Antic_PMG_issues.thumb.JPG.197805d1388ce7e18fceac3308937a56.JPG

 

I have a CO12296D ANTIC (from AMI) in my XLD and haven’t noticed any problems. Is there anything specific I should be looking for?

Link to comment
Share on other sites

1 hour ago, DrVenkman said:

I have a CO12296D ANTIC (from AMI) in my XLD and haven’t noticed any problems. Is there anything specific I should be looking for?

You'd know you had a problem if it were present. But basically there would be lines appearing and disappearing where there should not be anything. The best test for this is to go into the U1MB or Loader screens and use the up and down arrow keys.

 

So if your chip works, it could just be down to a batch and/or manufacturer variance that is enough to disturb the timing when tied into the other chips (i.e., GTIA, Sally) and/or whatever differences exist with the propagation delays through the glue logic.

 

  • Like 3
Link to comment
Share on other sites

26 minutes ago, shoestring said:

Like this.

But if you’re seeing it in an Incognito machine, that means it’s not a 1088XLD-specific design issue or anything @mytek should be specifically concerned about in his boards. You just have just run into a bad ANTIC independent of Michael’s concern.

 

Having said that, I have run into 2 - 3 bad ANTIC chips myself across my various A8/5200 machines, along with a few bad GTIAs, a bad SALLY or two, and a bad POKEY. These chips are 30 - 40 years old now so I guess we shouldn’t be too surprised.

  • Like 1
Link to comment
Share on other sites

2 hours ago, shoestring said:

Like this. It's not present in the loader screen though, at least not for me.

 

incognito.thumb.png.939623ba910ae8736716b32c296c664b.png

 

So I temporarily replaced the GTIA with a PAL version and thought I'd found the culprit. The replacement GTIA came but didn't help. I do have a spare ANTIC to try sometime tomorrow. Thanks for the heads up.

Yep that appears to be a PMG problem, although a bit different then what I was seeing. Based on my experience, and the fact that replacing the GTIA chip didn't help your situation, a new Antic would be the next logical thing to try.

 

Good luck :)

 

2 hours ago, DrVenkman said:

But if you’re seeing it in an Incognito machine, that means it’s not a 1088XLD-specific design issue or anything @mytek should be specifically concerned about in his boards. You just have just run into a bad ANTIC independent of Michael’s concern.

Good point, although I suspect that differences in hardware can contribute to making an Antic that works in one machine, fail in another. So don't throw away the chip without testing it in some other system where it could work perfectly fine ;) .

 

  • Like 2
Link to comment
Share on other sites

3 hours ago, DrVenkman said:

But if you’re seeing it in an Incognito machine, that means it’s not a 1088XLD-specific design issue or anything @mytek should be specifically concerned about in his boards. You just have just run into a bad ANTIC independent of Michael’s concern.

I suggested to shoestring that he post about his Incognito issue here because it seemed plausible that the two issues might be related, and the more information we have regarding individual cases, the easier it will be to draw conclusions about the exact cause of the problem. Since the XLD is built around U1MB and Incognito is essentially an U1MB for the 800 (and both use SRAM; Incognito uses a portion of the third 512K SRAM chip on the board for the base system RAM), perhaps the two problems are linked.

Edited by flashjazzcat
  • Like 2
  • Thanks 1
Link to comment
Share on other sites

1 hour ago, flashjazzcat said:

I suggested to shoestring that he post about his Incognito issue here because it seemed plausible that the two issues might be related, and the more information we have regarding individual cases, the easier it will be to draw conclusions about the exact cause of the problem. Since the XLD is built around U1MB and Incognito is essentially an U1MB for the 800 (and both use SRAM; Incognito uses a portion of the third 512K SRAM chip on the board for the base system RAM), perhaps the two problems are linked.

Yes I agree that these two situations might be related in some way, so it makes sense to discuss them here. Just how similar is the hardware responsible for the core functionality between the U1MB and the Incognito? Do they use the same CPLD, and if so, is the core logic arranged in the same way inside the chip for both cases?

 

Link to comment
Share on other sites

35 minutes ago, mytek said:

Do they use the same CPLD, and if so, is the core logic arranged in the same way inside the chip for both cases?

Same CPLD, yes (and same RAM, although the Incognito has 1.5MB instead of 1MB), and while I have not seen (and would not necessarily be able to fully comprehend) the JED sources for either device, I would be extremely surprised if a large proportion of the logic of one device was not re-used in the other. On the 6502 side, the firmware is 95 per cent identical and it is difficult or impossible for software to discern whether it is running on an U1MB or Incognito XL/XE mode. So - discounting the dual-personality nature of the Incognito - very closely related hardware.

  • Like 1
  • Thanks 1
Link to comment
Share on other sites

 

Well, I apologise if it looks as though I'm trying to derail a thread but I also felt the issues were somewhat related and could help someone else out. I've been literally pulling my hair out trying to find a cause.

 

Just an important note before you read on, both ANTICs are C012296D-01 ( replacement and original ).

 

Here is the original.

IMG_0484.thumb.jpg.74a6a8b053d821e2b414d20f1fa7b03a.jpg

 

With the replacement, the pmg issue was still present but only for a few seconds and completely disappears as the machine warms up ( maximum 10 seconds, average 5 seconds ). I can definitely live with that.  I'm no expert but I don't believe my ANTICs are faulty, there are just some subtle variances where the issue may be exhibited. I had the machine running for hours on end with the old ANTIC and the problem was permanent. Unlike the replacement,  the issue does not return as long as the machine isn't powered off for a significant amount of time then powered back on. 

 

 

 

Edit: Also, I forgot to mention that there was some PMG glitching at the left and right hand side of display when the screen saver kicked in. It's completely gone now with the new ANTIC.

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

13 hours ago, shoestring said:

Well, I apologise if it looks as though I'm trying to derail a thread but I also felt the issues were somewhat related and could help someone else out.

It's all good, and still points in the direction that not all Antics behave the same even when they're suppose to. This is probably especially true for the ones as old as what we get to use ;) . This could also simply be the case where the hardware application is getting close to exceeding the rated window of reliability timing-wise for the Antic. And just like a CPU, some can be overclocked, and some can't even when coming off the same silicon wafer.

 

Let's face it... BITD hardware was a lot simpler in some regards. Especially before a U1MB or an Incognito arrived on the scene. With something like the U1MB there is a lot of NEW logic in the path. So a big part of upgrading usually involves troubleshooting to some degree, and the inevitability that some chip swapping might be required.

 

Glad you got your system working correctly, aside from the minor warm-up issue :) .

 

  • Like 3
Link to comment
Share on other sites

Well I'll have to eat my words. When I went back to running a PAL GTIA and PAL Antic, I started seeing some issues after being left on for several hours, eventually resulting in a full lock up. I started thinking that this isn't right, since I had run these same chips for many, many hours in the recent past without any problems. So what changed?

 

Well I now had a fully ground planed board (RGB2VGA-XLD) floating above GTIA, Antic, and the CPU. Perhaps it was causing a heat problem, with the ground planed board reflecting it back down at the chips. I'd heard of Sallys sometimes exhibiting problems due to heat, and yes they do run quite warm in normal operation. Remembering that I had some stick-on 40-pin DIP heat sinks laying around, I opted to try one on my Sally.

 

P1220684.thumb.JPG.45847f12a7fb0454b52d19a9dd998046.JPG

 

It might be premature saying this, but the system has been running for over 4 hours now without issues :) .

 

  • Like 9
Link to comment
Share on other sites

10 hours ago, mytek said:

Remembering that I had some stick-on 40-pin DIP heat sinks laying around, I opted to try one on my Sally.

And I thought this was only necessary when Rapidus is installed in a machine alongside U1MB. :) Seriously: it may be that the problems I experienced with Rapidus (when it was still installed in a machine) are related; behaviour did after all change (in Rapidus' case, worsen) with increasing SALLY temps.

  • Like 4
Link to comment
Share on other sites

I left it running overnight in the U1MB setup screen and checked it 24 hours later. The screen was white and the system was locked up. So much for the heat sink curing the problem. Since this same CPU worked in the past, it's looking like something is beginning to fail internally. I can power it down and back up and all appears to work once again, but I'm sure if left running it would glitch again.

 

This CPU was purchased as NOS of which I bought two, so I swapped in its brother without a heat sink and have begun a long term test. Since I'll be gone for 5-6 days, it'll be a good test of reliability.

 

Link to comment
Share on other sites

4 days ago I replaced this exact same Sally chip for another as seen here in the photo.

 

Good_CPU.thumb.JPG.039ab7c4939191855f322626b4531918.JPG

 

No heat sink this time, and I ran the system with the top cover on, leaving it running for the entire 4 days uninterrupted.

 

5-day_run.thumb.JPG.75eedcdc6a47e49cd9aa81f3909cdaf2.JPG

 

After 4 days no visual aberrations or system lock-ups were seen.

 

So even though both CPU chips were from identical manufacturers, and both were NOS and bought at the same time from the same source, one started to fail no more than 6 months after its purchase. For those that haven't been keeping up with this thread, the failure was at first thought to be associated with a bad Antic chip while the 1088XLD was setup for NTSC mode, but then later similar issues arose when running a PAL setup as well. The problem at first appeared to be PMG related only, corrupting parts of the menu screens in the U1MB, but eventually got much worse causing all kinds of graphical glitches and ultimately resulted in a complete system lock-up.

 

So apparently even NOS chips can begin to fail with very little use. Since the Sally CPUs are not very expensive, it would make sense to buy 2 as I did, in case one becomes flakey.

 

  • Like 3
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...