WinUAE public beta #2

[Toni Wilen]

Read this before downloading or posting!

WinUAE public beta #2

Changelog: (mostly bug fixes)

- fixed possible crash when input recording stops (introduced in b1)
- log memory map after reset and also after autoconfig
- log loaded KS ROM information in memory map dump
- fixed boot crash (some CPU/KS combinations only)
- bug in new event handler fixed (symptom was random really slow non-ce/non-immediate blitter)
- interrupt priorities are correct again (broke in b1) Some tweaks.
- filesystem flag emulation updates (again) Protection flags on FAT volumes should now work properly in non-UAEFSDB mode.
- filesystem directory protection flags supported (on FAT volumes) Very old comment says: "Unix dirs behave differently than AmigaOS ones. windows dirs go where no dir has gone before..." Perhaps Windows 95 had some issues? Hope I didn't miss something..

[Paul]

I've just loaded HORIZON-VirtualIntelligenc demo, 3D Demo II and arte and so far so good. Speed problem looks fixed Thanks Toni

[Paul]

That reminds me, I don't know how long it's been like this but, I've noticed that burning spear demo 3CEDEBBD (in the zone) gets stuck when you see a blue circle with no rotating star. It works fine only when Cycle Exact mode is enabled.

[stainy]

Also fixed my crash dump thanks Toni

My sound issue is still there.. the little sound doowacky in the onscreen led`s thing goes right up to the 90`s and turns blue when it halts, then carries on..
Guess your gonna needs specs?

[rsn8887]

I just wanted to drop a quick note that I am still experiencing sound pitch shifts with 1.3.4b2.
They seem much less pronounced than in 1.3.3, but they are still very much audible.

At very low buffer settings (1 or 2), I still get scratchy sound, maybe because my machine is only 1.2 GHz - too slow? At higher buffer settings (3, 4 or 5) I get the very pronounced delay in sound fx. (e.g. fx after slashing sword in leander delayed by quite a long time (~500 ms?!) ), just as in 1.3.2.

However, in beta2 it is accompanied by occasional pitch shifts in music.

I am still waiting for my soundblaster live card to replace crappy VIA (god I hate these chipsets) onboard sound. I hope everything will be fine with the new card, if I ever get it (so much for ordering used stuff from amazon.com ). But I have the weird hunch that scratchy sound/delay will go away while pitch shifts in 1.3.3 and upwards will stay. But let's see.

For now, I still go with 1.3.2 for gaming. There I am sure that pitch is constant, while it might still be off a bit in general. I'd rather increase buffer, suffer the delay, but be certain to get a constant pitch.

[andreas]

Originally Posted by Paul

That reminds me, I don't know how long it's been like this but, I've noticed that burning spear demo 3CEDEBBD (in the zone) gets stuck when you see a blue circle with no rotating star. It works fine only when Cycle Exact mode is enabled.

I can't see anything wrong with this.
Cycle-Exact means "(68000) emulation as accurate as possible" and it's no big surprise some games and demos need this because they use very tricky coding and/or undocumented features.

[Toni Wilen]

Originally Posted by rsn8887

I just wanted to drop a quick note that I am still experiencing sound pitch shifts with 1.3.4b2.
They seem much less pronounced than in 1.3.3, but they are still very much audible.

Sound recorded to wave-file don't have any pitch problems, right?

At very low buffer settings (1 or 2), I still get scratchy sound, maybe because my machine is only 1.2 GHz - too slow? At higher buffer settings (3, 4 or 5) I get the very pronounced delay in sound fx. (e.g. fx after slashing sword in leander delayed by quite a long time (~500 ms?!) ), just as in 1.3.2.

This is caused by crappy sound card/driver, I remember seeing your log and your sound driver requested 5000+ bytes distance between play and write cursors -> long delay. Usual value is about 1000 or less.

I can reintroduce the lag adjustment setting (config file-only, no gui slider)

[spiff]

Not sure if this counts as a known problem, but battlesquadron running under A1200 config (either the AGA fixed version or WHD) causes corrupted graphics. Should I do a full bug report?

[Toni Wilen]

Originally Posted by spiff

Not sure if this counts as a known problem, but battlesquadron running under A1200 config (either the AGA fixed version or WHD) causes corrupted graphics. Should I do a full bug report?

Generally 68020+ game issues ("real" AGA game, "aga fixed" a500 game, whdload, whatever) will be ignored. CPU/customchip timing is not good enough and it won't be in 68020+ modes. (and <68020 = accuracy, 68020+ = speed)

(and offtopic in this thread anyway..)

[Crown]

I wih Toni could fix the Burning Spear demo as I have been taking part in its development and I wish I could run it with WinUAE.

[spiff]

CAPS image(s) hangs on boot, a500 quickstart. Works fine on 1.3.3


removed SPS image, do NOT post SPS images on EAB!! - Bippy

[spiff]

DMS image hangs on boot, works in 1.3.3 (prol related)

ADFs work

[Toni Wilen]

Originally Posted by spiff

CAPS image(s) hangs on boot, a500 quickstart. Works fine on 1.3.3

Fixed, another stupid bug in b1 (btw, it does not affect all caps images, only most custom loaders..)

I wih Toni could fix the Burning Spear demo as I have been taking part in its development and I wish I could run it with WinUAE

I thought this demo has "always" worked.

EDIT: And it still works just like on 1.3.3 -> OFFTOPIC HERE. (graphics crap during some parts = demo bug = ECS Agnus only)

[Toni Wilen]

Originally Posted by stainy

My sound issue is still there.. the little sound doowacky in the onscreen led`s thing goes right up to the 90`s and turns blue when it halts, then carries on..
Guess your gonna needs specs?

-sounddebug log needed and much more information as possible, like does it happen in every game? every configuration? buffer size?

[thor]

Automa (1991)(Vega).adf (CRC32 912556C1) has a graphic bug at the end. It happens with beta 1 and 2 and is not present when running it with 1.3.3. Configurations are the same as above.

[Toni Wilen]

Originally Posted by thor

Automa (1991)(Vega).adf (CRC32 912556C1) has a graphic bug at the end.

Argh, another stupid program starting new blit while previous has not yet finished (it is only a clearing blit and the area is already partialy cleared so programmers never noticed or something..)

In next beta both Hit Fido and Automa work (and hopefully without any more side-effects..)