HippoPlayer v2.48

[koobo]

Originally Posted by AmiClassic

I haven't used BlazeWCP for a long time.

This morning I was able to determine what was causing problem with scopes. Solution was to disable FBlit and CyberBugFix at the same time. Such a situation did not satisfy me, because I was losing about 700kb CHIP

Koobo has already solved this problem and in upcoming HiP version scopes work perfectly

Hippo was not compatible with this patch setup, it assumed to get CHIP memory for drawing whereas the patches resulted in FAST mem being delivered instead.

Doing custom blitter work instead of using standard gfx functions on such a buffer and expecting it to work is a false assumption. So now the memory type is detected, old blitter code is used for CHIP and new CPU based code is used for FAST. Possibly with FBlit and friends the scopes will now render faster than before as CHIP access is reduced Unless of course using an unaccelerated A500.

[daxb]

Originally Posted by DisasterIncarna

im the 1 with that patch and im not the one with the issue

We don't care about that fact. So, don't use patches that do the same like PatchControl and SaferPatches. Also check for obsolete patches with the list that comes with OS3.2? That might solve your non existing problems.

[DisasterIncarna]

Originally Posted by AmiClassic

I haven't used BlazeWCP for a long time.

This morning I was able to determine what was causing problem with scopes. Solution was to disable FBlit and CyberBugFix at the same time. Such a situation did not satisfy me, because I was losing about 700kb CHIP

Koobo has already solved this problem and in upcoming HiP version scopes work perfectly

out of interest, but doesnt FBlit have an exclusion option in its menus? maybe just exclude HippoPlayer instead of disabling ti altogether?

[AmiClassic]

Originally Posted by koobo

Possibly with FBlit and friends the scopes will now render faster than before...

Great to read such information

[daxb]

Yes, FBlit has include/exclude lists for FAllocBitMap. Here on OS 3.1 and the old HippoPlayer version HiP-Info, HiP-Scope and HippoPlayer are in the include list and the scopes work fine.

[DisasterIncarna]

Originally Posted by daxb

We don't care about that fact. So, don't use patches that do the same like PatchControl and SaferPatches. Also check for obsolete patches with the list that comes with OS3.2? That might solve your non existing problems.

cant remember where i read it, but i do recall "some" readme for "something" which suggested using

http://aminet.net/package/dev/debug/TRSaferPtch along with PatchControl (i think theres 2 SaferPatches, the linked good one and an obsolete/old one?), so er, i did, and nothing "seems" to have broken so kinda left it that way

[AmiClassic]

Originally Posted by DisasterIncarna

out of interest, but doesnt FBlit have an exclusion option in its menus? maybe just exclude HippoPlayer instead of disabling ti altogether?

This is something I haven't checked, but even if it worked, I would still have to turn off CBF losing mentioned CHIP memory. I preferred not to do that.

[koobo]

Originally Posted by daxb

Yes, FBlit has include/exclude lists for FAllocBitMap. Here on OS 3.1 and the old HippoPlayer version HiP-Info, HiP-Scope and HippoPlayer are in the include list and the scopes work fine.

I'm guessing that FBlit doesn't patch AllocRaster, maybe CyberBugFix does. I remember trying out FBlit as well with no problems (on A1200).
AllocBitmap is a kick3.0 function, AllocRaster is kick 1.2/1.3.

[AmiClassic]

I tried to check in Scout, but I still don't know anything

[koobo]

Originally Posted by AmiClassic

I tried to check in Scout, but I still don't know anything

Well, this is a mystery.
AllocRaster is besides very simple, it just does a basic memory allocation from CHIP, by default. So why would it return non-CHIP RAM if nobody has patched it...

EDIT: Does it being visible in the list mean that it is actually patched by someone? Ramlib is the system library loader task though.

[DisasterIncarna]

well, MCP is in that list and MCP has fastdraw/other patches that might be the culprit (ive pretty much got all gfx related patches in MCP disabled), could be worth disabling all of MCP first to see if 1 of its patches is doing it, if not thats 1 suspect down, if it is the culprit its just a case of disabling each gfx related patch until we have a winner.

[daxb]

Originally Posted by koobo

I'm guessing that FBlit doesn't patch AllocRaster, maybe CyberBugFix does. I remember trying out FBlit as well with no problems (on A1200).
AllocBitmap is a kick3.0 function, AllocRaster is kick 1.2/1.3.

CyberBugFix v1.1 (StingRay fix, 1600 bytes) seems to patch OpenLibrary() and AllocBitMap(). FBlit does patch several graphics.library functions (including AllocBitMap() ) but not AllocRaster().

EDIT: Does it being visible in the list mean that it is actually patched by someone? Ramlib is the system library loader task though.

I don't know about Ramlib but the answer is yes. For example FBlit and CyberBugFix are named in the Patcher row. So you know what task is responsible for the patch.

For the most users it would be the best/easiest to just start without Startup-Sequence and then launch the test program (in this case Hippo) to verify that no patch is responsible. Maybe run SetPatch and needed drivers first. If the problem/bug is gone you can start hunting patches or other software as a causer.

[peceha]

Do not go that way, just keep your code of the program frienfly with the system otherwise you will go crazy trying to find out what and how is being pattched
Let the oters follow the correct path

[AmiClassic]

I wouldn't be so critical. Reading thread 'Amiga OS 3.3 top 3 "must have" features (Wishlist)', I have a feeling that quite a few of these 'hardcore patches' might be part of system in near future

[tygre]

Hi Koobo and all!

I finally tested the latest HiP and I'm happy to report that it works very well, including MMD3 (like TwistedCinemaIII)

Siatek - Happy Bee 2 still doesn't play although another Impulse Tracker plays well (albeit with the usual "jumping screen").

MP2 files play choppy, maybe something to do with the preferences?

Cheers!

[tygre]

PS. On my config., playing any Impulse Tracker module make the screen "jump" and the screen will continue to jump with any other module, even in different format, until I restart HiP

[daxb]

I use the old Hippo version and have a question about a behaviour. If you skip forward in modules (pattern/position) usually the next pattern will be played until you reach the end of a module. For some module formats like XM the module ends already 3 patterns before the real end. E.g. an XM module has a length of 41 and when position 38 is playing a skip to 39 will end the module/plays next module in the list. What is the reason for this? Is it because of audio stream mixing buffer or something like this?

[koobo]

Originally Posted by daxb

I use the old Hippo version and have a question about a behaviour. If you skip forward in modules (pattern/position) usually the next pattern will be played until you reach the end of a module. For some module formats like XM the module ends already 3 patterns before the real end. E.g. an XM module has a length of 41 and when position 38 is playing a skip to 39 will end the module/plays next module in the list. What is the reason for this? Is it because of audio stream mixing buffer or something like this?

You're right that the actual playback is in the future compared to what is being heard. The bigger the buffer size and the lower the mixing frequency, the further in the future the playback will be. This may cause the module to end before you can actually hear it ending.

[koobo]

Originally Posted by tygre

PS. On my config., playing any Impulse Tracker module make the screen "jump" and the screen will continue to jump with any other module, even in different format, until I restart HiP

This is the same old strange behaviour. Is this correct:
- open HiP
- play Impulse tracker module
- boing!
- play protracker module
- boing again!
- close HiP
- open HiP
- play protracker module
- no boing?

[tygre]

Hi Koobo!

Originally Posted by koobo

This is the same old strange behaviour. Is this correct:
- open HiP
- play Impulse tracker module
- boing!
- play protracker module
- boing again!
- close HiP
- open HiP
- play protracker module
- no boing?

Yes, exactly

I'm using a Blizzard 1240 with an Indivision in an A1200...

Let me know if I could test other things to help you!