archive.worldofdragon.org

Fahfall uses single-bit sound for the footfall sounds.

I've found that the problem was dependent on the last value output when using non-single-bit sound. Fahrfall didn't trigger the (remaining!) bug, zephyr's diag cart rom did.

Ok, 0.29.3 released, I've done some measurements and adjusted various scales. Checked working with that diagnostic cart, fahrfall, demon attack and of course normal DAC use.

Thanks Ciaran!

Dragon 32 LLISTing is still not working in the latest Win32 version (xroar-0.29.3-w32). It stopped working somewhere between xroar-0.28.1-w32 and xroar-0.29-w32.

Alastair wrote:Dragon 32 LLISTing is still not working in the latest Win32 version (xroar-0.29.3-w32). It stopped working somewhere between xroar-0.28.1-w32 and xroar-0.29-w32.

Ah, the information that it ever did work is useful - I didn't realise. I'd only ever tested it in Dragon 64 mode. I'll give it a look over the weekend!

Fixed!

0.29.4 available with the fix, but I've also pushed out a snapshot build including many more development items here.

sixxie wrote:Fixed!

0.29.4 available with the fix, but I've also pushed out a snapshot build including many more development items.

A new release and a new snapshot? Ciaran you're spoiling us!

sixxie wrote:I've also pushed out a snapshot build including many more development items here.

It would be better to have a generic "-machine cocobr" (CoCo Brazil) which covers all four Brazilian PAL-M machines. I would also suggest continuing with the separate BASIC and Extended BASIC ROM format for the emulation of these Brazilian CoCo 1/2 clones. The MX-1600 is a clone of the later Korean CoCo 2 models which came fitted with a single 8K (For CoCo's fitted with Color BASIC 1.3 only) or 16K ROM chip in the same 28 pin socket. Earlier US CoCo 2 models always came fitted with two separate 8K ROM chips like the CoCo 1.

I love the emulated 6847T1 VDG , but think it should be disabled by default because it was never fitted to CoCo 1 models, and not all B model CoCo 2's come with this chip fitted as standard.

I'm still having problems LLISTing a program and I've discovered that it is not the Dragon 32 emulation that is at fault. If I start XRoar 0.29.4 or snap-20130223 with "xroar -machine dragon32 -lp-file test.txt" I find that LLIST works, but if I use "xroar -lp-file test.txt" and CTRL M to the Dragon 32 machine then LLIST stops working. A further test of pressing CTRL M all the way through back to the Dragon 64 (PAL) machine also causes LLIST to stop working. So it seems likely that at least one of the other emulated machines is causing the LLIST print-to-file problem.

By using the GUI of XRoar-0.29.4-gtk-w32 to select machines I've found that the Tano Dragon 64 (NTSC) emulation is to blame. The Tano emulation will LLIST properly but selecting another Dragon machine after emulating the Tano causes LLIST to fail.