I did not check that many, but it did happen with the KQ2 booter. DDPG booter in tandy failed to start for me as well, but does work properly in PCjr. KQ1 booter works OK in PCjr. I can try more when I get some time. One other thing I noticed is the screen capture crashes DOSBox. The following capture of KQ2 was done with the print screen.

That is what I saw. The KQ1 PCjr. booter has an issue with the text not clearing itself.

Confirmed! All I need to do is set the machine type to pcjr and boot the game. Now here's some interesting facts... Found another regression. You only get the green dots in the Tandy version only. Text remains on the screen but you can still play the game. Also it doesn't crash for me when I try to take a screenshot for King's Quest 1.

Now here's the weird part. In the IBM PCjr version of the game I don't see and dots. The text is still on the screen and it's not refreshed. But on top of that I can't play the game. You can type stuff in but you cannot move! I'm pressing the arrows but it doesn't work just like in 0.74.

Edit: Tried King's Quest 2 and it doesn't crash for me... I can use the screenshot capture feature.

I was using the EmuCR build, if that makes any differences.

collector wrote:

I was using the EmuCR build, if that makes any differences.

Yes indeed! Seems like the EmuCR builds crash every single time you try to snap a screenshot. I think that build is compiled with Visual Studio 2010... so is that another bug? MinGW and cross compiled versions work fine and it doesn't crash when you press Ctrl+F5.

Mine is compiled with VS2010 and doesn't crash with Ctrl-F5...

Aren't the EmuCR builds done without PNG capture support?

http://www.dosbox.com/wiki/SVN_Builds#Plain.2 … lean_SVN_builds

Compiled without screenshot, video recording and IPX support.

Checked only with the KQ1 PCJR booter (the others aren't available on my harddisk). The issue of text not being cleared was introduced with r3815.

The attached patch seems to fix it for PCJr. However it can't fix anything for Tandy. The latter probably has a similar implementation as PCJr.

The patch does fix the text refresh on KQ1, but KQ2 and BC still show the gray and green dots.

New patch attached. Clearing video memory on SetMode was also affected. What is BC?

So we have several problems:
- screenshots, probably caused by the build (works fine for me)
- KQ1 booter in PCJR mode, ScrollWindow scrolling incorrect region of memory - fixed in the patch
- KQ2 booter and other showing dots after SetMode, display memory cannot be cleared through the B800 segment completely on the JR - fixed in the patch

- KQ2 booter locking up hard with dynamic core - known issue or new?
- DDPG booter not starting up with machine=tandy (occurs in both r3850 and r3812), works in 0.74

h-a-l-9000 wrote:

What is BC?

The Black Cauldron

The updated patch solves the problem with the green dots.

The KQ2 booter needing normal core is not new. The DDPG issue may have something to do with an error in the image itself, as it fails to boot in my real Tandy 1000SX or PCjr. (128KB) but boots fine in my PC.

The regression in DDPG is related to r3626. A result of Retrograde Station's crack that locates its INT 13h handler as high as possible in conventional memory based on the amount of memory reported by INT 12h. The handler is being placed in graphics memory in SVN, so it gets wiped out when the switch to graphics mode occurs.

My understanding about DDPG is not so much of a crack as there are three versions floating around. Two of them are fan ported to PC, one from Atari and one from Amiga. At least that is what AGINFO IDs them as. Both of these will crash when you hit Escape. The third is a booter image. I have two Tandy versions that I cannot read the disks.

My understanding is that only DDPG booters were being discussed here... In any case, my previous comment is in regard to the booter from Retrograde Station.

I wasn't trying to challenge you or anything. I am just trying to sort out DDPG. I had not heard of it having copy protection. The only reason that I mentioned the other two is that this game seems to have several different versions, each with its own issues. Was the copy protection on the booter CPC?

Pressing ctrl-v in the Retrograde Station booter reports "Version 1.0Q - June 9, 1986 - Adventure Game Interpreter Version 2.001". Oversized sectors are part of the protection, hence the INT 13h handler set high up in memory to deal with those, so it may be similar/related to CPC if not exactly that.

I usually just invoke debug (Alt+D) in the AGI games to get game and interpreter versions, but yeah, the booter should be 1.0Q according to HWM's Sierra interpreter list.