Unicorns season: Prince of Persia for the A8!

[2Tom]

Now I can't check how it works on real hardware.

 

Maybe this option is responsible for this error?

 

[_The Doctor__]

Never seen it happen on real hardware...

[rensoup]

10 hours ago, 2Tom said:

Hi

In V1.01 Final Release I sometimes get this error after starting the game, I tested the game on the Altirra 3.90 and Atari800 4.2.0 emulator.

 

 

Thanks, I got it to happen too using those same steps, pressing the button during the text screen after the cutscene... I'm going to investigate.

 

@MrFish reported it a while back too...

 

On 11/8/2021 at 4:29 AM, MrFish said:

I got this error when starting my first game today, using the Atarimax cart. It went away after moving beyond the first screen.

 

My configuration is the same as before.

Altirra v3.90 (release)

600XL/800XL

OS Rev. 2

 

The only option I used for the game was "No" flickering fonts.

I wasn't able to repeat the error yet. So, I'm not sure if it had anything to do with when I started the game (what part of the startup sequence). I know I didn't watch all of the intro.

 

I forgot to get a savestate file too.

 

 

glitch.wmv 4.26 MB · 37 downloads

 

 

[rensoup]

10 hours ago, 2Tom said:

Now I can't check how it works on real hardware.

 

Maybe this option is responsible for this error?

I believe I have a fix... don't know when then next release will be though.

 

10 hours ago, _The Doctor__ said:

Never seen it happen on real hardware...

It should happen on real HW, could anybody try ? it's quite easy to reproduce... just follow 2Tom's vid

[+x=usr(1536)]

On 12/10/2021 at 6:46 PM, rensoup said:

It should happen on real HW, could anybody try ? it's quite easy to reproduce... just follow 2Tom's vid

Here ya go

 

Test environment: NTSC 800XL, UAV, 256K RAM, SIDE3 (current firmware) / FujiNet (current firmware), CX78 joypad.

 

Cartridge image testing:

 

Intro Font Flicker set to 'Y' (default):

• Atarimax: no issues
• Megacart: no issues
• SIC: no issues

 

Intro Font Flicker set to 'N':

• Atarimax: no issues
• Megacart: first-screen corruption present and matches previous screenshot / video
• SIC: first-screen corruption present and matches previous screenshot / video

 

Disk image testing:

 

Intro Font Flicker set to 'Y' (default):

• Single-Density: no issues
• Double-Density: no issues

 

Intro Font Flicker set to 'N':

• Single-Density: no issues
• Double-density: first-screen corruption present and matches previous screenshot / video

 

Ran through the tests three times each and the results were consistent: with Intro Font Flicker set to 'N', the Megacart, SIC, and Double-Density images all displayed the first-screen corruption.

 

One thing that might be of value would be to test this on a completely stock 128K 130XE: my thinking is that this could effectively eliminate any memory upgrades as being a potential problem.  I don't have any real reason to believe that they are, but it would still be good to at least have that testing as a point of reference.

[rensoup]

6 hours ago, x=usr(1536) said:

One thing that might be of value would be to test this on a completely stock 128K 130XE: my thinking is that this could effectively eliminate any memory upgrades as being a potential problem.  I don't have any real reason to believe that they are, but it would still be good to at least have that testing as a point of reference

Thanks for the extensive testing!

 

(You could have stopped after encounter the first corruption though ?)

 

My guess is that it would happen with a stock 130XE because it's a somewhat simple mistake from my part, though I don't know why it doesn't seem to happen with the other textscreens (if my fix is right...)

[+x=usr(1536)]

36 minutes ago, rensoup said:

Thanks for the extensive testing!

 

(You could have stopped after encounter the first corruption though ?)

No worries - and I spent a few years early in my career working in QA.  Some old habits die hard

36 minutes ago, rensoup said:

My guess is that it would happen with a stock 130XE because it's a somewhat simple mistake from my part, though I don't know why it doesn't seem to happen with the other textscreens (if my fix is right...)

Completely understood, and agreed.  I was thinking it would be a quick way of eliminating memory upgrades on 64K (and lower) machines, but I am honestly not convinced that that is the underlying issue.  Think of it as more of a nice-to-check

[+-^CrossBow^-]

14 hours ago, rensoup said:

My guess is that it would happen with a stock 130XE because it's a somewhat simple mistake from my part, though I don't know why it doesn't seem to happen with the other textscreens (if my fix is right...)

I'm pretty sure I've seen it come up at once on my memory stock 130xe using my FujiNet to load the game. I've not tried the newest release but happy to give it a go.

 

[VinsCool]

Look what i just received in the mail!

 

[Atari8guy]

3 minutes ago, VinsCool said:

Look what i just received in the mail!

 

I'm jealous

 

[+-^CrossBow^-]

Wanted to mention that this past weekend I did try and get the same corruption to occur on my 130xe changing the font flicker to N and was NOT able to get it to happen. This was with the latest 1.01 release of the game through my FujiNet. But... I also just remembered that from the first time I encountered that glitch, to now I did replace all of the RAM on my 130xe as mine was still using the MT ram. Well, all but the first bank which uses an NEC brand. I now have the NEC in the first socket and Samsung brand in the others all running at the same speed unlike how it was originally. 

 

So memory could still play a part in this and I changed the conditions with my RAM replacement.

 

[+x=usr(1536)]

53 minutes ago, -^CrossBow^- said:

Wanted to mention that this past weekend I did try and get the same corruption to occur on my 130xe changing the font flicker to N and was NOT able to get it to happen. This was with the latest 1.01 release of the game through my FujiNet. But... I also just remembered that from the first time I encountered that glitch, to now I did replace all of the RAM on my 130xe as mine was still using the MT ram. Well, all but the first bank which uses an NEC brand. I now have the NEC in the first socket and Samsung brand in the others all running at the same speed unlike how it was originally. 

 

So memory could still play a part in this and I changed the conditions with my RAM replacement.

 

That's actually a good point, and I should check on what's in the 800XL.  I'm fairly certain that whoever did the 256K upgrade on it didn't use MT RAM for the upgrade, but they may have left the original memory in there as the lower 64K.

 

Late edit: checking photos in the following thread from a few months back, they certainly did leave the original MT RAM in there.

 

 

It definitely hasn't been changed out since then, so is still in there.  Hmm.  Might be worth swapping out and seeing what happens.

[Level42]

Finally had the chance to download V1.01 and I can confirm that sometimes it happens on my 600XL with U1MB using SIDE3 too, but definitely not always.

[Level42]

15 hours ago, x=usr(1536) said:

That's actually a good point, and I should check on what's in the 800XL.  I'm fairly certain that whoever did the 256K upgrade on it didn't use MT RAM for the upgrade, but they may have left the original memory in there as the lower 64K.

 

Late edit: checking photos in the following thread from a few months back, they certainly did leave the original MT RAM in there.

 

 

It definitely hasn't been changed out since then, so is still in there.  Hmm.  Might be worth swapping out and seeing what happens.

Not ALL mT RAMS are bad. The 130XE that I bought way back in the mid 80's has each and every RAM mT and is running just fine until this day.
I am pretty sure it was a certain batch of week numbers of production that was bad. Atari seemed to know about it because almost always the top left position on the 130XE was always another brand.

[+x=usr(1536)]

1 hour ago, Level42 said:

Not ALL mT RAMS are bad. The 130XE that I bought way back in the mid 80's has each and every RAM mT and is running just fine until this day.

Ditto the 800XL.  This is more just to satisfy my curiosity.

1 hour ago, Level42 said:

I am pretty sure it was a certain batch of week numbers of production that was bad. Atari seemed to know about it because almost always the top left position on the 130XE was always another brand.

Entirely likely.  It's also possible that MT was throwing second-quality chips in with 100% good ones on large orders and letting it be the downstream manufacturer's problem.

[Level42]

21 minutes ago, x=usr(1536) said:

Ditto the 800XL.  This is more just to satisfy my curiosity.

Entirely likely.  It's also possible that MT was throwing second-quality chips in with 100% good ones on large orders and letting it be the downstream manufacturer's problem.

Maybe, but I can totally imagine the thought of one of the Tramiels accepting an offer for a very good price on "almost good" RAMs

[+-^CrossBow^-]

2 hours ago, Level42 said:

Not ALL mT RAMS are bad. The 130XE that I bought way back in the mid 80's has each and every RAM mT and is running just fine until this day.
I am pretty sure it was a certain batch of week numbers of production that was bad. Atari seemed to know about it because almost always the top left position on the 130XE was always another brand.

Yes... as I stated the upper left RAM in my 130xe was actually an NEC branded one at 250ns but the rest of them were MT running at 300ns. I replaced them out with Samsung 250ns a few weeks back although to be fair, the originals hadn't shown any issues that I'm aware of. But I like to bullet proof stuff whenever I can.

 

[xxl]

1 hour ago, Level42 said:

Maybe, but I can totally imagine the thought of one of the Tramiels accepting an offer for a very good price on "almost good" RAMs

what's weird about that? today you accept almost working U1MB in the field of D3xx and nobody sees a problem in it

[+DjayBee]

Who said Yehova?

[+Stephen]

1 hour ago, xxl said:

what's weird about that? today you accept almost working U1MB in the field of D3xx and nobody sees a problem in it

Crazy isn't it?  Some people also take a barely working single speed loader when we have HDDs that are literally 60 times faster.  People are just crazy I tell ya!

[Peri Noid]

5 hours ago, xxl said:

what's weird about that? today you accept almost working U1MB in the field of D3xx and nobody sees a problem in it

'Cause it's not a problem. 

[+Stephen]

4 hours ago, Peri Noid said:

9 hours ago, xxl said:

what's weird about that? today you accept almost working U1MB in the field of D3xx and nobody sees a problem in it

'Cause it's not a problem. 

Hey - there's two pieces of software that won't work with this "junk" upgrade.  Some obscure program from the early 1980s that was hand picked and noone heard of, and a game from xxl that he wrote on purpose to not work on U1MB (and was subsequently fixed by patching 2 bytes of code).  It's obviously a major problem!

[tep392]

31 minutes ago, Stephen said:

Hey - there's two pieces of software that won't work with this "junk" upgrade.  Some obscure program from the early 1980s that was hand picked and noone heard of, and a game from xxl that he wrote on purpose to not work on U1MB (and was subsequently fixed by patching 2 bytes of code).  It's obviously a major problem!

I’ve been using my U1MB for two years and never knew it had a problem until now. ?‍♂️

[Marius]

17 hours ago, xxl said:

what's weird about that? today you accept almost working U1MB in the field of D3xx and nobody sees a problem in it

 

Do you know what is weird... that under your avatar on the left I read "Content moderated - indefinitely" and your comment did make it to the thread.

Look, let me be very clear. Some time ago I decided to work from now on with stock atari 8bit only. This was a good decision -FOR ME!
I focus mainly on external solutions. I have my Turbo Freezer attached and my IDE+ 2.0 and from time to time I hook up my MyIDE 1 or 2 interface. All cool stuff. 

But does this mean that other people here on AtariAge need to follow that same route? Hell no! Can I decide for other people what they should use? Of course not. U1MB is one of the best and generally accepted internal a8 upgrades ever created. @flashjazzcat did an amazing job writing the best possible firmware and thanks to him U1MB became even triple as powerful (or even more). I am not as enthusiastic about everything Candle invented, but his U1MB is awesome. 

In all the years I worked with U1MB I noticed one serious problem and that was with the US Speed Copy v5 by E. Reuss. On U1MB it turned out it wasn't reset proof. Was this a fault of U1MB. No. Was this a fault of US Speed Copy v5? Hmmm neither. The two did not get along due to sharing same memory position in stack. It was very easy for me to patch the US Speed Copy v5, problem solved. Deciding to change the architecture of the Atari 8bit computer by adding something to it, inevitably means a compromise. Always. Not every program likes to be booted from a High Speed Sio drive. Not every program plays along nicely with a Stereo Upgrade. Some programs did only run fine on a real 130XE and not on a 256K 800XL. 

Compared to that, the problems with U1MB are virtually zero. You have proven your point that it is possible to write something that does not work. But that is not so difficult. In stead of proving and defending this retarded and obnoxious point of view, you'd better use your skills, knowledge and humanity, to make the atari 8bit scene better and friendlier. 

The reason I took out my U1MB has only one reason. I like to move from Atari 8bit to Atari 8bit and I did not have installed U1MB in all of them. So I chose the 'freedom' to focus mainly on Stock atari 8bit. But that had nothing to do with a lack of quality. U1MB is awesome, and I might even decide to buy the external U1MB as soon as it is on the market. 

[derSammler]

12 minutes ago, Marius said:

Do you know what is weird... that under your avatar on the left I read "Content moderated - indefinitely" and your comment did make it to the thread.

That makes you weird. Moderation is not about censorship. While you may not like his comment, it's nothing a moderator should reject in the first place. I don't even see where's the issue with what he wrote. Maybe some people just take written words too seriously.