ZeroPage Homebrew Twitch Stream

[+D Train]

rage resetting and masking appropriately on the subway:

 

 

my camera makes things look all fuzzy!  weird

[+ZeroPage Homebrew]

44 minutes ago, D Train said:

rage resetting and masking appropriately on the subway:

my camera makes things look all fuzzy!  weird

Love it, the pic is going on the show today! I just got mine last week so I'll be showing it off today on the stream too. ? 

 

- James

[+sramirez2008]

1 hour ago, D Train said:

rage resetting and masking appropriately on the subway:

 

 

my camera makes things look all fuzzy!  weird

I assume this picture was taken while traveling on an actual "D" subway Train ?

[+ZeroPage Homebrew]

The exclusive world premiere from @cd-w we played today has been revealed as Zeviouz!! We had a blast playing it and you will be blown away by the skill that has gone into making this look so good and play so well! You'll forget you're playing it on an Atari 2600, it's that good, you have to try it!

 

Head over to the Zeviouz thread and DOWNLOAD IT NOW!!

 

 

[+D Train]

6 hours ago, ZeroPage Homebrew said:

Love it, the pic is going on the show today! I just got mine last week so I'll be showing it off today on the stream too. ? 

 

- James

glad that you got it in time!  it was fun seeing you in your shirt and my picture up there.  We are such dorks.

 

for anyone wondering, that is the large mask option and it's not bad!  the mask itself is two layers, sewn at the top and bottom so that you can stick a filter between them.  it's a little big for me, but the quality is better than I thought it would be.

 

5 hours ago, sramirez2008 said:

I assume this picture was taken while traveling on an actual "D" subway Train ?

that was the 2 train.  I live off the brighton line (which was formerly the D but is now the Q), so when I go into work, I take the Q from the house, transfer to the 2 or 3 in times square, and take that up to harlem where my office is.

 

when I lived around the corner from yankee stadium I took the D all. the. time.  it is in fact, why I am the D train!

[+sramirez2008]

1 hour ago, D Train said:

that was the 2 train.  I live off the brighton line (which was formerly the D but is now the Q), so when I go into work, I take the Q from the house, transfer to the 2 or 3 in times square, and take that up to harlem where my office is.

 

when I lived around the corner from yankee stadium I took the D all. the. time.  it is in fact, why I am the D train!

Nice...but the 2 train was my line. I lived across from the Gun Hill Rd. station (#2 line) in the Bronx.  While I don't have any pics of myself on the train, I do have a pic of the station. Btw, Zeviouz is amazing!!! 

[scrummy]

It was beautiful and surprising.  That moment on the stream when we all saw the rotating phantom zone slabs rotating down the screen.  Congratulations to all involved.  This is a nice development. 

[Thomas Jentzsch]

What happened in Frantic at 2:02:14? Bug?

[+ZeroPage Homebrew]

1 hour ago, Thomas Jentzsch said:

What happened in Frantic at 2:02:14? Bug?

Pesco has revealed that my TIA apparently suffers stray PF pixels. ?  Would a TIA chip replacement solve the issue?

 

- James

 

[Thomas Jentzsch]

Maybe. Might be worth a try.

[+ZeroPage Homebrew]

12 minutes ago, ZeroPage Homebrew said:

Pesco has revealed that my TIA apparently suffers stray PF pixels. ? 

Update, my machine has gone full Pesco, there's PF pixels all the way down the left. ?

 

EDIT: I read up on the issue and they didn't show up in my previous screenshot because I had already eaten the corresponding dots in the third from the right columns.

 

I'll be conscious of this when playing games on the stream to point out that it's my system and not the game. This is coupled with the colour timing issue that's introduced with the RGB mod.

 

- James

 

Edited August 19, 2020 by ZeroPage Homebrew

[+Nathan Strum]

3 hours ago, ZeroPage Homebrew said:

Pesco has revealed that my TIA apparently suffers stray PF pixels. ?  Would a TIA chip replacement solve the issue?

3 hours ago, Thomas Jentzsch said:

Maybe. Might be worth a try.

 

If you want to give it a shot, it should be an easy enough swap. Just get yourself a proper chip lifter to carefully pry the TIA out. It's a very snug fit in the mod board (you shouldn't have to remove anything else - just the TIA). If you have another 2600 around with a socketed TIA, just use the chip from that one. But test it in the other console first to make sure it doesn't have the same problem.

 

I use this to remove chips: https://www.amazon.com/dp/B00433SJB2/

 

Just carefully work the lifter under the chip, gradually prying it up evenly a little from each side. Too bad we didn't know about the warmup issue at the time I redid the mod - I could've swapped the TIA easily then. Live and learn. (On the plus side, now I know something else to look out for when repairing my 2600. And John's.  )

[+johnnywc]

5 minutes ago, Nathan Strum said:

 (On the plus side, now I know something else to look out for when repairing my 2600. And John's.  )

Soon to be THREE consoles (2 2600's and a 7800).  Dislikes: Lightning ? 

[+ZeroPage Homebrew]

1 hour ago, Nathan Strum said:

If you want to give it a shot, it should be an easy enough swap. Just get yourself a proper chip lifter to carefully pry the TIA out. It's a very snug fit in the mod board (you shouldn't have to remove anything else - just the TIA). If you have another 2600 around with a socketed TIA, just use the chip from that one. But test it in the other console first to make sure it doesn't have the same problem.

 

I use this to remove chips: https://www.amazon.com/dp/B00433SJB2/

 

Just carefully work the lifter under the chip, gradually prying it up evenly a little from each side. Too bad we didn't know about the warmup issue at the time I redid the mod - I could've swapped the TIA easily then. Live and learn. (On the plus side, now I know something else to look out for when repairing my 2600. And John's.  )

Thanks for the link! I'll check out my other 2600 systems to see if any of the chips are socketed and also pass the Pesco test.

 

I wonder if anybody has made an all-in-one piece of software for issues with the 2600. Tests for all the various input devices, chip defects, video/audio issues, color timing, etc. That would be super useful!

 

- James

[Omegamatrix]

6 hours ago, ZeroPage Homebrew said:

Update, my machine has gone full Pesco, there's PF pixels all the way down the left. ?

 

EDIT: I read up on the issue and they didn't show up in my previous screenshot because I had already eaten the corresponding dots in the third from the right columns.

 

I'll be conscious of this when playing games on the stream to point out that it's my system and not the game. This is coupled with the colour timing issue that's introduced with the RGB mod.

 

- James

 

I realize you're thinking the mod is the culprit, because the issue only appeared after the mod.

 

However, I did take a look in Stella using you screenshot. I can see that left side of playfield is being updated 1 cycle late (it ends on cycle 38 instead of 37). This dot that you are seeing is one of the pixels from how PF2 was last set on the previous line, but the rest do get overwritten before displayed so just the one dot appears out of place.

 

You may see this on other systems that don't have a RGB mod, because the write timing is out of spec. Here is a table of the times provided by SeaGtGruff.

 

 

 

 

 

 

[+Nathan Strum]

43 minutes ago, ZeroPage Homebrew said:

I wonder if anybody has made an all-in-one piece of software for issues with the 2600. Tests for all the various input devices, chip defects, video/audio issues, color timing, etc. That would be super useful!

I've had a speculative project in my to-do list for years called the 2600 Toolbox, which would basically do all that. I have a half-written blog entry about it that remains unpublished. But I've been thinking about it a lot lately in light of things like testing consoles for bus stuffing compatibility, TIA glitches, color calibration, AtariVox testing, and on and on.

 

The downside is, it requires people who can actually program to make it happen.  

[+ZeroPage Homebrew]

THIS FRIDAY ZeroPage Homebrew has the EXCLUSIVE WORLD PREMIERE of Leandro Camara's (@HSW3ET) new Atari 2600 homebrew Deep Space! 

 

You may know Leandro's previous 2600 games A New Maurader and Rally Racer released through More Work Games and now he's back with a brand new homebrew that takes us back into the (space) battle!

 

"On board of the NIX 2-6 Spaceship you have been informed that four Motherships are transporting stellar matters from the mines of a near planet. Your mission is to intercept the Motherships and recover the content, but before you need to browse the four orbits where each Mothership is and face the challenges of DEEP SPACE."

 

We'll be debuting DEEP SPACE this Friday on ZeroPage Homebrew's Twitch livestream at 12PM PT | 3PM ET | 7PM GMT, make sure you tune in!

 

Twitch Stream: https://www.twitch.tv/zeropagehomebrew/

 

[+stephena]

11 hours ago, Omegamatrix said:

However, I did take a look in Stella using you screenshot. I can see that left side of playfield is being updated 1 cycle late (it ends on cycle 38 instead of 37). This dot that you are seeing is one of the pixels from how PF2 was last set on the previous line, but the rest do get overwritten before displayed so just the one dot appears out of place.

 

You may see this on other systems that don't have a RGB mod, because the write timing is out of spec. Here is a table of the times provided by SeaGtGruff.

 

Incidentally, there's a setting in the developer options in Stella to force this 'Pesco' behaviour.  It's under Options -> Developer -> TIA -> Developer Settings -> "Glitched Pesco".  It's listed as being caused by "Delayed playfield bits".

[+ZeroPage Homebrew]

On 8/19/2020 at 7:29 PM, Omegamatrix said:

I realize you're thinking the mod is the culprit, because the issue only appeared after the mod.

 

However, I did take a look in Stella using you screenshot. I can see that left side of playfield is being updated 1 cycle late (it ends on cycle 38 instead of 37). This dot that you are seeing is one of the pixels from how PF2 was last set on the previous line, but the rest do get overwritten before displayed so just the one dot appears out of place.

 

You may see this on other systems that don't have a RGB mod, because the write timing is out of spec. Here is a table of the times provided by SeaGtGruff.

That's an extremely useful table for programmers, thanks for the link. I'll be looking into replacing my TIA with a known good one to fix the timing issue on the rare occasion that a game like Pesco will exhibit this behaviour.

 

14 hours ago, stephena said:

Incidentally, there's a setting in the developer options in Stella to force this 'Pesco' behaviour.  It's under Options -> Developer -> TIA -> Developer Settings -> "Glitched Pesco".  It's listed as being caused by "Delayed playfield bits".

That's an incredibly handy option in Stella so programmers can try to avoid causing the Pesco behaviour in misbehaving TIA chips! It's definitely better to try to work around these types of problems while making your game if possible. Thanks!

 

- James

Edited August 21, 2020 by ZeroPage Homebrew

[RevEng]

On 8/20/2020 at 1:21 PM, ZeroPage Homebrew said:

I'll be looking into replacing my TIA with a known good one to fix the timing issue on the rare occasion that a game like Pesco will exhibit this behaviour.

I understand where you're coming from - wanting to have a show go as smoothly as possible - but I'm not as sure replacing your TIA is a win overall. A percentage of legit TIAs are just like yours. So the fault lies with game code that doesn't run correctly on some subset of real hardware. FWIW, the same glitch appeared during your show when you featured Space Cactus Canyon, and it led to my fixing the timing in the bB DPC+ kernel. 

 

I know you've started down the road of 2600 development; I'd personally hang tightly onto such a corner-case TIA. For similar reasons, I haven't disabled the infamous dark chambers "timing-fix circuit" on my 7800.

[+Nathan Strum]

1 hour ago, RevEng said:

I understand where you're coming from - wanting to have a show go as smoothly as possible - but I'm not as sure replacing your TIA is a win overall. A percentage of legit TIAs are just like yours. So the fault lies with game code that doesn't run correctly on some subset of real hardware. FWIW, the same glitch appeared during your show when you featured Space Cactus Canyon, and it led to my fixing the timing in the bB DPC+ kernel.

I think that's an excellent point. Sure, Pesco may always look wonky, but that's no real great loss.  And having the glitch appear on the show also led Darrell to address it in Frantic.

[Thomas Jentzsch]

I think you are right. Developers should use Stella's developer options and check how their game handles on multiple TIA variants.

[+sramirez2008]

Just now, Thomas Jentzsch said:

Stella's developer options and check how their game handles on multiple TIA variants.

I don’t use Stella that often, so I wasn’t aware that it emulates TIA variants. Wow! So I can play with or without “Pesco” behavior. You guys have thought of everything.?

[+ZeroPage Homebrew]

2 hours ago, RevEng said:

I understand where you're coming from - wanting to have a show go as smoothly as possible - but I'm not as sure replacing your TIA is a win overall. A percentage of legit TIAs are just like yours. So the fault lies with game code that doesn't run correctly on some subset of real hardware. FWIW, the same glitch appeared during your show when you featured Space Cactus Canyon, and it led to my fixing the timing in the bB DPC+ kernel. 

 

I know you've started down the road of 2600 development; I'd personally hang tightly onto such a corner-case TIA. For similar reasons, I haven't disabled the infamous dark chambers "timing-fix circuit" on my 7800.

I was discussing it with Darcy on yesterday's stream and it's really not worth replacing the TIA for such rare occasions that the problem occurs. And to your point, it helps during development to make sure that all games work on all systems properly, even ones with defects! The glitch on my system has now lead to improvements on two occasions, so it's doing more good than harm.

 

A similar issue is going on with BUS development to also insure that all games will work on the lowest common denominator of systems.

 

My 2600 game is using asymmetric playfields so I will definitely have to watch the timing of the PF. So you're right, this issue may crop up for me as well and having a system like mine to test it on will be a great help in diagnosing issues!

 

- James

 

 

[Mr SQL]

1 hour ago, ZeroPage Homebrew said:

I was discussing it with Darcy on yesterday's stream and it's really not worth replacing the TIA for such rare occasions that the problem occurs. And to your point, it helps during development to make sure that all games work on all systems properly, even ones with defects! The glitch on my system has now lead to improvements on two occasions, so it's doing more good than harm.

 

A similar issue is going on with BUS development to also insure that all games will work on the lowest common denominator of systems.

 

My 2600 game is using asymmetric playfields so I will definitely have to watch the timing of the PF. So you're right, this issue may crop up for me as well and having a system like mine to test it on will be a great help in diagnosing issues!

 

- James

 

 

 

James it may not be the TIA that is causing random playfield pixels to be set - I had a 2014 build of KC Munchkin where this was happening but it would only happen after the game was on over night. I found it was temperature dependent as an Atari in a cold room could trigger it in just a couple of hours.

 

Turns out some of the low Zero Page memory locations can change with temperature (not the 128 bytes of RAM) and if you are using as an offset, the offset value changes causing the stray playfield pixels to be set. 

 

What's stranger is that the phenomena only happened with my Harmony running KC as per the description from the vid -  

This strange behaviour showed up only after a 20 hour burn in on my Harmony running KC; 48 hour plus burn ins on SuperCharger run clean.

 

@batari any thoughts if these are related or what is causing them?

   

Edited August 22, 2020 by Mr SQL
post finished automatically on lookup