Jump to content

Photo

Today's Delivery


25 replies to this topic

#1 SmittyB ONLINE  

SmittyB

    Moonsweeper

  • 287 posts

Posted Thu Sep 13, 2018 2:30 PM

So I got a delivery today and I'm now the very happy owner of an AtariVox and Seagull 78 adaptor.

Naturally the first thing I did was to answer a question that has been on my mind ever since I saw the Seagull 78, and yes... It does indeed work flawlessly with the Sega Mega Drive's infrared controllers. I did use an extension cable because while it would probably plug in directly I didn't want to force it. Now I can play wirelessly... with only 8 feet of cable plugged in... while I sit right next to 7800 to use the console switches... (I've decided to stick to the wired pad for now). I had a quick couple of games of Xenophobe and Xevious, and it's surprisingly noticeable how much better I am than when using the standard 'europad'. I had to remember that I could use both buttons at the same time with Xevious because the recessed and spongy buttons on the europad make that completely impractical.

 

For the AtariVox I scraped a little of the plastic out of the sides of the connector on one of my extension cables so I can leave it in the housing. I thought I'd have to seriously hack apart the socket to make it fit but I just shaved a little of the plastic off with a penknife and it fits nicely. I'm enjoying hearing it chime out "Atari vox!" when it powers up. Naturally I had to test it so I gave Dungeon Stalker a whirl and loved every minute of it.

I'm definitely going to have to play around with this and see what I can make it do.

 

The world has been conspiring against me to keep me from getting these so I'm happy to finally get my hands on them.



#2 -^CrožBow^- ONLINE  

-^CrožBow^-

    Quadrunner

  • 6,491 posts
  • Collector of Fine Atari and Sega Antiquities!
  • Location:Ivory Tower, Fantasia (Oklahoma)

Posted Thu Sep 13, 2018 2:34 PM

On Dungeon Stalker, does your Avox go crazy and stop working when the wizard shows up? Cause it does on both of my 7800s and on pretty much all the others I've tested it on.



#3 SmittyB ONLINE  

SmittyB

    Moonsweeper

  • Topic Starter
  • 287 posts

Posted Thu Sep 13, 2018 2:45 PM

I don't think so. I'd better play another few rounds... for testing purposes of course.



#4 SmittyB ONLINE  

SmittyB

    Moonsweeper

  • Topic Starter
  • 287 posts

Posted Thu Sep 13, 2018 2:58 PM

So after some thorough testing, I've come to the conclusion that I'm very bad at this game, and that no it hasn't failed for me though I only just about reach the wizard and then die horribly without being able to get a hit on it.

I am in PAL-Land so it could just be the difference in timing on my machine.


Edited by SmittyB, Thu Sep 13, 2018 3:02 PM.


#5 SmittyB ONLINE  

SmittyB

    Moonsweeper

  • Topic Starter
  • 287 posts

Posted Thu Sep 13, 2018 3:45 PM

I tell a lie, I just played again and had it corrupted when the Wizard and I died at the same time while it was still saying whatever it says when the Wizard appears, so trying to say 3 phrases at a time. I would still get 'voice' but it was more like a string of phonemes with a constantly rising inflection as if all further data was offset in some way. When the next phase would have played I got a similar sounding noise.

#6 RevEng OFFLINE  

RevEng

    River Patroller

  • 4,994 posts
  • Bitnik
  • Location:bottom of the stack

Posted Thu Sep 13, 2018 5:06 PM

I tell a lie, I just played again and had it corrupted when the Wizard and I died at the same time while it was still saying whatever it says when the Wizard appears, so trying to say 3 phrases at a time. I would still get 'voice' but it was more like a string of phonemes with a constantly rising inflection as if all further data was offset in some way. When the next phase would have played I got a similar sounding noise.

 
This is depressing. There were three testers with AtariVox during the original Dungeon Stalker development, and none of us can reproduce it. I've had independent reports of others reproducing it, and others that can't. And the ones who can reproduce it seem to be able to do it on every 7800 they own. I was sent a Vox from someone who did reproduce the issue, but it ran clean on my 7800.
 
I don't believe this is a software bug, based on the testing done in this thread. Also, T:ME Salvo intentionally used less speech, compared to Dungeon Stalker, but people have reported the bug is worse there. I think that may be due to more required stick handling.

I've considered ripping the AtariVox speech support out in 7800basic. I'm not sure what else to do.

Sorry to rain on your happy thread. It's a very cool device, 7800 voice issue notwithstanding.

#7 -^CrožBow^- ONLINE  

-^CrožBow^-

    Quadrunner

  • 6,491 posts
  • Collector of Fine Atari and Sega Antiquities!
  • Location:Ivory Tower, Fantasia (Oklahoma)

Posted Thu Sep 13, 2018 5:30 PM

Dungeon stalker is the only game I've found to have an issue with my Avox+. It works normally otherwise for any other game that features speech with it. 

 

Last we left with this when testing, was to try using a Harmony encore with the testing program to see if could replicate it. But again, I've tested DS with my Avox+ on both of my 7800s and on a few that I've modded for others and it always resets and stops working shortly after the wizard shows up. Could it just be a difference between the original Avox and these newer ones?



#8 RevEng OFFLINE  

RevEng

    River Patroller

  • 4,994 posts
  • Bitnik
  • Location:bottom of the stack

Posted Thu Sep 13, 2018 5:47 PM

Dungeon stalker is the only game I've found to have an issue with my Avox+. It works normally otherwise for any other game that features speech with it.

I've had reports of Salvo being worse, for those that have the issue with Dungeon Stalker. It has nothing to do with the Wizard round itself - during various attempts we disabled some and all of the Wizard speech, and the vox crash would just be less frequent but still show up during non-Wizard gameplay.

In the tests we did with your gear, we found that 2600 mode was a lot more robust. I'm pretty sure this is why you don't have issues with 2600 vox games. That, and the fact that most 2600 titles don't get close to stuffing the buffer, so they don't rely on timely "buffer full" advice.

The low level driver in 7800basic is the same one (from Alex Herbert) used in the 2600 games. The only change was timing, and this was ruled out with testing.

 

Last we left with this when testing, was to try using a Harmony encore with the testing program to see if could replicate it. But again, I've tested DS with my Avox+ on both of my 7800s and on a few that I've modded for others and it always resets and stops working shortly after the wizard shows up. Could it just be a difference between the original Avox and these newer ones?

I don't think it's necessarily the vox units. I had a newer one sent to me, and I still can't reproduce the problem. The ability to reproduce or not-reproduce seems to solidly stay with the person, rather than the gear, but I don't know what that fact actually points to.

The only data point to the contrary is jinks - he reported that he had the issue with a regular cart, but didn't run into the issue on Concerto.

#9 Trebor OFFLINE  

Trebor

    River Patroller

  • 4,624 posts

Posted Thu Sep 13, 2018 7:12 PM

Five consoles, four different serial lines, including RF plain and modded (Include simple, "Best Electronics", and 2 Longhorn) and no problems with either of the two AtariVox modules owned (Both being later "+" edition) with either Dungeon Stalker or T:ME Salvo.
 
I don't believe ripping Vox support out of 7800basic is necessary; however, perhaps a mention in the 7800basic guide about possible sound issues have been reported by some users should be added.  To state the least, the (high score) saving support option should remain.
 
Pardon my absent mindedness, but did we come to any sort of consensus surrounding power (consumption)/adapters being a possible contributing cause? 

 

Particularly in the case of video mods, is a lack of power affecting timing(s)/performance?  Are some mods perhaps consuming some additional power that others are not, that typically would not be a noteworthy factor, outside of an item such as the AtariVox?  Or is it not any mods at all, and maybe the AtariVox performs a certain power draw or power condition that would normally not be present otherwise (?)

 

When users are testing multiple consoles, is the same power supply being used with just a swap of the console itself?



#10 RevEng OFFLINE  

RevEng

    River Patroller

  • 4,994 posts
  • Bitnik
  • Location:bottom of the stack

Posted Thu Sep 13, 2018 7:23 PM

Pardon my absent mindedness, but did we come to any sort of consensus surrounding power (consumption)/adapters being a possible contributing cause? 
 
Particularly in the case of video mods, is a lack of power affecting timing(s)/performance?  Are some mods perhaps consuming some additional power that others are not, that typically would not be a noteworthy factor, outside of an item such as the AtariVox?  Or is it not any mods at all, and maybe the AtariVox performs a certain power draw or power condition that would normally not be present otherwise (?)
 
When users are testing multiple consoles, is the same power supply being used with just a swap of the console itself?


I did ask for power supply swaps in one case, and that didn't make a difference. -^Cro§Bow^- added a power conditioning capacitor and new regulator to one of his 7800s, with no positive impact.

In my most tinfoil-hat'ish moments I'm suspecting low or dirty power from the power company, or large amounts of EMI, being a factor.

#11 Jinks OFFLINE  

Jinks

    River Patroller

  • 4,435 posts
  • Location:Canada

Posted Thu Sep 13, 2018 8:18 PM

Other than these 2 7800 games are there any other avox games that crash the avox when used on a 7800?
Fall Down and Juno First all work ok.

What about extension cables used? Drop of voltage thru the possibly cheap wiring?
Or even pin tension on the contact points of the extension cable or avox pin thickness?

#12 RevEng OFFLINE  

RevEng

    River Patroller

  • 4,994 posts
  • Bitnik
  • Location:bottom of the stack

Posted Thu Sep 13, 2018 8:38 PM

Other than these 2 7800 games are there any other avox games that crash the avox when used on a 7800?
Fall Down and Juno First all work ok.

What about extension cables used? Drop of voltage thru the possibly cheap wiring?
Or even pin tension on the contact points of the extension cable or avox pin thickness?


There are no other games that use vox voice in 7800 mode. The tests -^Cro§Bow^- ran replicated the issue on in 2600 mode if he was aggressive with the triggering (and only with the joystick movement, not with button), but it seemed way more robust than the 7800 mode tests.

The cabling doesn't seem to be a factor. I asked someone with the issue to test a direct plug-in (IIRC with a naked system) and the issue still occurred.

#13 -^CrožBow^- ONLINE  

-^CrožBow^-

    Quadrunner

  • 6,491 posts
  • Collector of Fine Atari and Sega Antiquities!
  • Location:Ivory Tower, Fantasia (Oklahoma)

Posted Thu Sep 13, 2018 9:44 PM

All of the testing I did was with the Avox casing off and plugged directly into my port 2 on my console. I also have two actual 7800 power supplies and neither made a difference. Also as Rev stated, I actually have a DC-DC 1amp output converter in my primary 7800, but this testing I believe was done prior to my installing that into my 7800. I've played and tested the game since and it still has these issues.

 

My other 7800 has been very slightly modded in that it was setup using one of Eckard's DevOS setups. Otherwise it is still completely stock as far as components and only having RF output on it. Other 7800s I tried with the game did it both unmodded and after modding. I use Dungeon Stalker for burn in testing since it has a good demo mode and with the different colors of the playfield mazes it makes for good checks for color bleed etc after modding.

 

One of the 7800s I tested with had both the DC-DC converter installed and I had added a 2.5x5mm jack to allow use of a Genesis model 1 power supply and it still did it with that supply as well.

 

The only other power related issue I can think of, is that the power to my house is high..it averages at about 125v. (My kill-a-watt has recorded between 123-127v). I also have one of those new fangled digital meters that the power company installed about 2 years ago. That stupid thing generates RF noise that I can pick up with older RF output consoles..ugh... I suspect it is the wireless comms it uses to phone home and allow them to get their readings and remotely control the power to peoples houses without actual site visits required anymore.

 

But yeah, the fact that I could replicate similar results with different 2600s as well really only leaves on consistent factor in my testing...

 

The Avox+ itself...



#14 -^CrožBow^- ONLINE  

-^CrožBow^-

    Quadrunner

  • 6,491 posts
  • Collector of Fine Atari and Sega Antiquities!
  • Location:Ivory Tower, Fantasia (Oklahoma)

Posted Thu Sep 13, 2018 9:51 PM

Five consoles, four different serial lines, including RF plain and modded (Include simple, "Best Electronics", and 2 Longhorn) and no problems with either of the two AtariVox modules owned (Both being later "+" edition) with either Dungeon Stalker or T:ME Salvo.
 
I don't believe ripping Vox support out of 7800basic is necessary; however, perhaps a mention in the 7800basic guide about possible sound issues have been reported by some users should be added.  To state the least, the (high score) saving support option should remain.
 
Pardon my absent mindedness, but did we come to any sort of consensus surrounding power (consumption)/adapters being a possible contributing cause? 

 

Particularly in the case of video mods, is a lack of power affecting timing(s)/performance?  Are some mods perhaps consuming some additional power that others are not, that typically would not be a noteworthy factor, outside of an item such as the AtariVox?  Or is it not any mods at all, and maybe the AtariVox performs a certain power draw or power condition that would normally not be present otherwise (?)

 

When users are testing multiple consoles, is the same power supply being used with just a swap of the console itself?

 

Terbor I believe the general concensus on what is happening is that the input buffer on the Avox+ is getting hammered faster than it can drain. As a result we end up with a timing issue in that the Avox+ essentially seems to run out of ram space before it has drained off the last command. As a result it only gets part of what it needs and that is what causes the gibberish and resetting of the device. Rev is correct in that it didn't take much using his test to cause issues in the 7800. And in the 2600, I was never able to get it to mess up just using the fire button to activate. Only when using the joystick controller to execute the phrases in the test rom would I eventually get the 2600 to cause the same issue.

 

Juno 1st I believe is the only game I have for the Atari systems that uses its speech and that has never been an issue.

 

I don't recall if I tested T:ME Salvo though...I think I have a later non-final rom image of that I could pop into the Mateos to check this weekend perhaps?



#15 RevEng OFFLINE  

RevEng

    River Patroller

  • 4,994 posts
  • Bitnik
  • Location:bottom of the stack

Posted Thu Sep 13, 2018 10:10 PM

The buffer is getting stuffed, but the buffer-full line is either not signalling consistently or not in time.

On my unit, when I stuff the buffer, it signals full. Writes are then deferred until it drains.

#16 SmittyB ONLINE  

SmittyB

    Moonsweeper

  • Topic Starter
  • 287 posts

Posted Fri Sep 14, 2018 1:05 AM

I had a read the of the SpeakJet documentation and I can see mention of the 'buffer-half-full' signal but no mention of what is expected to happen when the buffer overflows. Am I wrong in thinking that unless you check for that signal before writing each byte it would be quite easy to get close to that limit then write more than 32 bytes in one go and overflow?

#17 RevEng OFFLINE  

RevEng

    River Patroller

  • 4,994 posts
  • Bitnik
  • Location:bottom of the stack

Posted Fri Sep 14, 2018 6:34 AM

Yes, the line needs to be checked, or else you can easily overflow. The 7800basic driver does this.

#18 SmittyB ONLINE  

SmittyB

    Moonsweeper

  • Topic Starter
  • 287 posts

Posted Fri Sep 14, 2018 6:38 AM

I'll see if I can rig up a test to see how much I can send in one go before running in to problems

#19 SmittyB ONLINE  

SmittyB

    Moonsweeper

  • Topic Starter
  • 287 posts

Posted Fri Sep 14, 2018 11:24 AM

Just an idea... I'm not really a hardware guy but part of the problem I had getting keypad controllers to work in my game was due to how you have to wait for capacitors linked to the controller ports to charge on the console before reading it back and how in the end I had to practically double the 400ns wait described in the programming spec.
Perhaps something along those lines is delaying the signal just enough to cause the problem. Would explain why it affects different consoles in different ways.

#20 RevEng OFFLINE  

RevEng

    River Patroller

  • 4,994 posts
  • Bitnik
  • Location:bottom of the stack

Posted Fri Sep 14, 2018 12:19 PM

Yeah, it's entirely possible something along those lines is why avox voice in 7800 mode is less robust than 2600 mode.

I wanted to make a clarification, since I was posting from my cell last night - the proper name of the line I was referring to as "buffer-full" was in fact "D0/Ready". This is the line AtariVox uses to check if device isn't ready (ie. buffer full, chip is initializing)

The half-full line you referenced isn't wired to, or used by, AtariVox.

#21 SmittyB ONLINE  

SmittyB

    Moonsweeper

  • Topic Starter
  • 287 posts

Posted Fri Sep 14, 2018 1:16 PM

Hmmm. I modified my keypad test program to have a variable delay and actually my 7800 appears to be on spec. The keypad requires a 400 microsecond delay and it starts to function correctly between 120-150 cycles (342 - 428 microseconds). Of course I only experienced the voice problem in dungeon stalker when I triggered at least 3 phrases at the same time so maybe being on spec is why.
 
I've attached the test for people to try. Keypad in second port naturally, joypad in first port to change the delay. I failed to correct a bug where the delay shown is 1 higher than what's used due to the order of checks but hey, it gets the point across.
 
If my hypothesis is correct then the more susceptible to the 7800 is to the voice issue the higher a delay it will require to read the keypad correctly.
 
The right side of the pad will tend to work fine for the most part, as you increase the delay you'll start to get more accurate readings, but as you press multiple keys you'll get brief false readings flash up. When you get the correct display all the time is when you'll want to make a note of your readings and compare it against another machine.


Just had the issue again. It doesn't seem to be as severe on my machine in that the avox doesn't reset. I fought the Wizard once with no problems, then the next time after moving around the screen a few times it said something like "I have advanced" followed by a brief pause, then "aaaaaaaaaaaaaaaaaaaaaaaaah", then he took off in his roflcopter which goes "tuh tuh tuh tuh tuh tuh tuh".

Attached Files


Edited by SmittyB, Fri Sep 14, 2018 1:33 PM.


#22 RevEng OFFLINE  

RevEng

    River Patroller

  • 4,994 posts
  • Bitnik
  • Location:bottom of the stack

Posted Fri Sep 14, 2018 2:59 PM

I don't have a keypad, but I think someone else here does.

It's a longshot request, but do any of you that can reproduce the issue have access to a UPS?

#23 SmittyB ONLINE  

SmittyB

    Moonsweeper

  • Topic Starter
  • 287 posts

Posted Fri Sep 14, 2018 3:06 PM

It's a longshot request, but do any of you that can reproduce the issue have access to a UPS?


I do not

#24 Trebor OFFLINE  

Trebor

    River Patroller

  • 4,624 posts

Posted Fri Sep 14, 2018 4:02 PM

Keypad (Video Touch Pad) I do own.  A UPS I do not own, unfortunately.

 

*EDIT:  And I just realized the other check is reproducing the issue, in which I also do not.  :dunce:



#25 -^CrožBow^- ONLINE  

-^CrožBow^-

    Quadrunner

  • 6,491 posts
  • Collector of Fine Atari and Sega Antiquities!
  • Location:Ivory Tower, Fantasia (Oklahoma)

Posted Fri Sep 14, 2018 6:48 PM

I do have a video touchpad if I can locate it LOL! I have a set of the keypad controllers for 2600 Basic Programming...that would work as well.

 

Let me give it a shot in a bit...






0 user(s) are browsing this forum

0 members, 0 guests, 0 anonymous users