Enhancer Bug thread

@xenic

Glad that I could help.

It is too bad that the sound card issues were not caught during the X5000 beta testing.

@Amigakit

As "Updater" seems to be part of Enhancer now...

I get an error message of a missing AmiDVD 1.51 on the server side, please fix

Calendar started from the menu "execute command" throw a DSI like this:

Crash log for task "calendar"
Generated by GrimReaper 53.19
Crash occured in module calendar at address 0x7D40A040
Type of crash: DSI (Data Storage Interrupt) exception
Alert number: 0x80000003

Register dump:
GPR (General Purpose Registers):
0: 80000000 6232E150 00000000 00000000 7FFA2B3C 0196029C 0196029C 00000000
8: 688B92A0 00000000 00000044 0181EA28 42842888 63B24BD0 00000000 652ED340
16: 7D40D4A4 00000000 63BFE760 63C19848 02270000 02270000 00000000 00000001
24: 6FF93180 00000001 46842888 00000000 6FF93180 01848CFC 63C19828 6232E150


FPR (Floating Point Registers, NaN = Not a Number):
0: nan 255.5 0.99999 0
4: 0 0 0 0
8: 0 -0.00196464 3.8598e-06 -0.00393181
12: 1 1 0 0
16: 0 0 0 0
20: 0 0 0 0
24: 0 0 0 0
28: 0 0 0 0

FPSCR (Floating Point Status and Control Register): 0x82024000


SPRs (Special Purpose Registers):
Machine State (msr) : 0x0200F030
Condition (cr) : 0x620652CC
Instruction Pointer (ip) : 0x7D40A040
Xtended Exception (xer) : 0x6415B018
Count (ctr) : 0x6FF694D0
Link (lr) : 0x7FB22920
DSI Status (dsisr) : 0x6415ADD0
Data Address (dar) : 0x01A95E28



680x0 emulated registers:
DATA: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
ADDR: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
FPU0: 0 0 0 0
FPU4: 0 0 0 0



Symbol info:
Instruction pointer 0x7D40A040 belongs to module "calendar" (HUNK/Kickstart)

Stack trace:
module calendar at 0x7D40A040 (section 5 @ 0x2101C)
module calendar at 0x7D40A028 (section 5 @ 0x21004)
native kernel module newlib.library.kmod+0x000020a4
native kernel module newlib.library.kmod+0x00002d0c
native kernel module newlib.library.kmod+0x00002ee8
calendar:_start()+0x170 (section 1 @ 0x16C)
native kernel module dos.library.kmod+0x000255c8
native kernel module kernel+0x00042618
native kernel module kernel+0x00042660

@lazi

Ooops, sorry, just fixed ;)

Thank you for the report

Hello,

It seems that there is an issue with Wave datatypes, even with 1.3 release.
Some wave files couldn't be loaded any longer.

I've built an exe from a SDK sample.
You can dowload it here:
http://boing.attitude.online.fr/PlayAHI.lha

Source and wave files included.

A few users told me that it doesn't work on their X1000, X5000.

It works well on my "non enhanced" AmigaOS 4 machines (and these wave files can be loaded on MorphOS & Win too).

Any test & help would be appreciated as it prevents sounds from being played by both Ask Me Up XXL and Word Me Up.

Thank you.

@glames
I reported problems with sound datatypes not returning correct attributes with GetDTAttrs(). Read pages 6 & 7 of this topic for details. The sounds will be played by datatypes, you just can't get the correct attributes. The datatypes have been fixed and I assume the fixes will show up in the next Enhancer release.

You could just play your sounds with datatypes instead of getting attributes with datatypes and then playing then with AHI.

@Xenic,

Thank you for info.

I need to play sounds with AHI in my games. So a quick fix would be very appreciated.

And please run the sample exe I sent before the "Enhanced pack" next release to check it works well after new fixes.

@glames

I don't know how to launch them with your PlayAHI as i got an error

Error load sample son_debut.wav
Error load sample testDT.wav

But if i double click directly on the wav files they will be played correctly in MultiViewer ...

I'm using wav.datatype 53.28 from the latest Enancher pack beta release

@glames
I had the same results as samo79. Even though the improved Enhancer sound datatypes I tested have fixed the issues with sound attributes for SDTA_SampleLength, SDTA_SamplesPerSecond & SDTA_BitsPerSample; the SDTA_Sample attribute is still broken. Your sound program won't even work with the improved datatypes because getting the SDTA_Sample attribute is returning 0 instead of the sample address. With the current public release, your program probably isn't getting any valid sound attributes.

Here is a quick review of problems I found with Enhancer sound datatypes:
1. Not sending soundplay termination signal - Fixed
2. Not returning correct attribute SDTA_SampleLength - Fixed
3. Not returning correct attribute SDTA_SamplesPerSecond -Fixed
4. Not returning correct attribute SDTA_BitsPerSample -Fixed
5. Not returning correct attribute SDTA_Sample - Still broken
6. Plays mono sample in left channel instead of both channels
like OS4.1u1 sound.datatype - Still broken

How can we expect 3rd party programmers to write working software when system components are being replaced with broken ones. If a commercial product is going to replace system components the replacements should work at least as well as the original components. That's just my opinion.

@Samo79,
You run it well.

When it works, it displays:
playing 5842 bytes at 11025 hz
playing 2147 bytes at 11025 hz

And two sounds are played :)

As Xenic explained, the datatype is now loaded but you can't extract l all data. IDataTypes->GetDTAttrs doesn't work as it should and I guess MultiViewer doesn't use it.

@Xenic,

After your previous replay, Iasked a user to test it again and indeed, SDTA_SampleLength , SDTA_SamplesPerSecond and SDTA_BitsPerSample are now OK but not SDTA_Sample that returns zero :(

Of course, I fully agree. Updated system components like datatype were not enough tested. I reported the bug late January and it still doesn't work six months later...

It is really bad. My customers and even resellers that bundle my games complains about it... And it's normal.

So now, let's go forward and let's try to be constructive.

IMHO, there should be a list of software/games that should be tested before every "Enhancer" release. Of course, I think about my games/software (Dir Me Up, Ask Me Up XXL, Word Me Up) but not only. Entwickler-X, Cherry Darling, ... and others products should be tested too, even free ones like for example Odyssey.

If this list had existed, the datatype issue would have been detected before release.

@Xenic @Glames

I have passed on your reports, thank you.

Please understand we have always responded quickly to user and developer reports and have a track record of resolving any issues that come to light. So far we have released 4 major public updates of Enhancer Software in just over a year and all of our users can have confidence that we will continue to improve the software through Updater and future releases.


Regarding the new Sound Datatype from the Enhancer Software:

As per the documentation, Getting the SDTA_[Left|Right]Sample attributes when streaming sound results in a null response (that's from V44 autodocs). The A-Eon datatypes stream by default.

@amigakit
Quote:
That doesn't explain why SDTA_Sample is returning a NULL attribute for 8SVX and AIFF soundfiles since those would be loaded by the OS4.1FE 8SVX and AIFF datatypes. I was going to suggest to glames that he switch to 8SVX or AIFF sound files but that won't help either at this point.

@glames
Quote:
In order to avoid future problems like this, I can only suggest that you convert your sound files to AIFF and get libaiff from OS4depot so you can load the files and get the attributes for AHI without relying on datatypes.

@mattew
Probably now you can see why i all the time says that replacing any os4 components are bad idea: originals already was tested by more bigger users base years ago and bug fixed till state they are now. Making replacement now, with much smaller user base we have now, will returns in the years of bugfixing and developrs/users problems till its all will be the same as originals => which is time, which can be spend on anything else which we didn't have.

@kas1e

100% ACK!

@glames
Quote:
While that would help, a better idea would be for developers to have a set of automated tests. Warp3D Nova's automated test suite has helped me catch a lot of stuff before release. When someone does find a bug I missed, then a new test is added.

Writing automated tests can be a pain (it takes extra time), but it pays off in the long run.

Hans