Quote:
I'm replying to myself because I have resolved my problems with Dialer/AmiFTP. As usual it was the a1serial.device causing the problems. No a1serial.device since OS4 pre3 has worked properly with my ?A1. I reported the problem over and over and even ranted a little but to no good end. I don't think anyone is testing a1serial.device with a ?A1 on a dialup connection. The main symptom is always the same: large files can't be uploaded and large attachments can't be sent with an email using YAM. I've reverted to a1serial.device v51.11 (12/26/2005) and all is well now with Dialer/AmiFTP.

@Skov
Quote:

I'm also experiencing problems with a dialup connection.

If I've been online for more than a few minutes and select "Disconnect" in the Dialer window:
Dialer window won't execute my shutdown script that plays a notification sound and won't close the window. It will iconify but won't close.

AmiFTP won't upload large files to my WEB space. It send's about 25k and then just sits there. If I select "Abort" it takes a couple of minutes for AmiFTP to throw up an error requester but that may just be a timeout issue. After the upload failure, Dialer definately won't close it's window when I select "Disconnect".

When I reset my?A1 and boot with my OS 4.0 boot CD, AmiFTP will upload large files and Dialer works properly.

I'm having problems with several Internet applications. Dialer is acting strangely. If I have been online and accessing WEB pages for more than about 10 minutes, selecting "Disconnect" in the Dialer window has no effect for about 20 seconds. After about 20-30 seconds my modem disconnects but the Dialer window remains open. It should disconnect immediately, execute a script that plays a "shutdown" notification sound and close the window. The Dialer window will iconify/deiconify but can't be closed.

I was attempting to upload some files to my WEB space with AmiFTP but could not upload files larger than about 25k. AmiFTP just stopped and did nothing. I was trying to upload some images for links in the AW.net thread "My impression of Amiga OS 4.1" to show Rogue what is happening in IBrowse with compositing on my ?A1. When I booted with OS4.0 (using an OS4.0 boot CD) both Dialer and AmiFTP worked normally.
I attempted to upload images using AmiFTP with OS 4.1 several times to confirm that the problem was reproduceable.

The Dialer disconnect problem can be reproduced immediately if I attempt to send a large file with AmiFTP, abort the upload after AmiFTP stops, wait for AmiFTP to display an error requester and then select "Disconnect" in the Dialer window. It will take about 20 secs for the Modem to disconnect and the Dialer window then won't close.

@PEB
I bought it at the link AJS posted. It's a Download and you get it immediately with a credit card. Only downside is the HTML docs which refer you to the "manual" in a number of cases. I'm new to image processing and it's taking a lot of trial-and-error to figure out how to do some things with Photogenics. Wish I had the book
P.S. Photogenics5 opens a WorkBench cloned screen on it's own and looks really nice with OS 4.1.

@yescop
A TurboCalc upgrade patch from Aminet.

@Swoop
Copy this file too:
SYS:Prefs/Env-Archive/Sys/screens.prefs
My Screens settings from OS4 worked.

@saimo
Quote:

That should have been user configurable in ASL prefs. Maybe it will be added in an update. Dopus4 works that way so I'm used to double-clicking.

@saimo

Quote:
That should have been user configurable in the ASL prefs. Maybe it will be added in an update. It doesn't bother me too much because Dopus4 works that way.

@yescop
What version of Turbocalc are you using. I have v 5.02E and it seems to be working. I did notice that if some prefs items are set to things not available in a fresh OS 4.1 install (like fonts) Turbocalc fails to work properly.

@Murakami
Quote:
I don't see a "Force" option in OS4 ASL prefs so I assume that it's another new addition that I will have to disable. I'm glad PageStream 4 works with OS 4.1 I wish my copy would arrive. It was sent 5 days ago. It must be on a slow flight over the Atlantic ocean.

@Murakami

Quote:

Since PageStream 4 seems to clone WorkBench by default maybe some of the new features like "compositing" are a problem. Have you tried turning off that feature? Have you tried changing the ASL prefs? My copy of OS 4.1 is still on the way but I'm going to be really pissed if PageStream doesn't work. Over the years I've spent far more for PageStream upgrades than I did for OS 4.1. I can't believe that OS 4.1 would not have been beta tested with a major application like PageStream so there must be a solution.

@Ricossa

Quote:

When Deron switched to MUI for PageStream 5, I didn't bother to upgrade because it is unlikely that MUI will ever be updated on OS4.xx. PageStream, BME & Pageliner all open on OS4 style screens without using the "Screens" prefs. I just select "New Public" in Pagestream/Screen Preferences and I get an attractive OS4 screen (like WorkBench).

@Snuffy


Quote:

I emailed Software Hut at the beginning of July and when I received no response, I too ordered from AmigaKit. However, I just received a response from SoftHut yesterday and Joe wasn't even aware of the OS 4.1 release. My interpretation of his email is that he was expecting to hear from Amiga Inc. about any new developments for A1/OS4 (laughing loudly at the thought that AI is doing anything useful). I can't really blame Joe for concentrating on earning a living when the Amiga market is barely alive in the U.S.

@LiveForIt

Quote:

Indeed there should be. I don't know which "New stuff" you are referring to but it should be fixed so it won't crash on OS4. I hope you reported the problem to the author.

@klesterjr

Quote:
At least you can edit your system icons. I find it really irritating that the OS4 System Icon Editor can be used to edit the current System Icons. Before I upgraded to ?A1/OS4 I used to spend hours drawing/modifying icons to suit my tastes. Now I just have to accept somebody elses icons as they are. It sucks. How are they drawing these new system icons? How can I modify them?

I'm really pissed off about the new OS4 programs that force you to install TBImages and don't seem to give you any way to replace them with your own. I really had fun creating my own images for YAM, IBrowse etc. to replace some of the ones I didn't like. If you're going to give us some new type of icons for Amiga, give us a way to draw/edit them as well. Otherwise, I'm not really interrested.

@number6

Strangely, I just got the same timeout when I clicked an
AmiKit link in another forum here. Is the Amiga world
under attack or something??

@salass00

I just have a few minor comments.

The version reported for the DT descriptors is:
"J2K . (29.7.08)" & "JP2 . (29.7.08)"
They need to be fixed so the version command interprets them correctly. (That way the version of the descriptors and the datatype will match)


The source files for the descriptors (J2K,JP2) start with the comment:
# DataType file for a RIFF-WAVE Sound

You might want to change RIFF-WAVE to JPEG2000 to avoid any confusion for developers.


Could you place the next upload in a directory before you archive it so the files don't get mixed with other files that might already exist in the destination dir?

Thanks for the good work.

P.S. Why do we need 2 descriptors (J2K, JP2)??

@Snuffy

Relax. I wasn't trying to mount a personal attack on you. I just thought your statement was a little misleading and that your advice to delete those files wasn't sound. RAD won't survive a reboot with OS4 Final but it still works and can be quite useful. For example, I downloaded the ImageMaster files just released by Blackbelt. They are released as ADF files or DMS files. I downloaded the ADF files and used the "Transadf" program to write the seven adf files to 7 880k RAD drives which I mounted as RAD1, RAD2 etc. with the unit numbers changed to 0 through 6. I clicked on the disk1 installer and installed ImageMaster to my hard-disk. I haven't tested the program yet but the install worked. Granted, there are probably quicker and easier ways to accomplish the installation but I just did it as an exercise to demonstrate the flexibility and usefullness of the old RAD drive.

The older Amiga documention refers to the Recoverable RAM Drive (RAD) as Ramb0: but only as RAD: in the OS 3.1 documentation. Before OS4 Final was released the RAD drive would survive an Amiga reboot (Ctrl RAmiga/LAmiga) but not a warm reset. I used a 64 MB RAD drive with OS4 pre4 and did most of my compiling on the RAD drive because it was faster than compiling on my hard-disk. If the Grim Reaper popped up when I tested a program I could simply reboot and all my source files remained in the RAD drive. RESRAD is too small to use for compiling large programs and slows down a reset because it clears all the memory it uses to 00 when started.

I want my mountable recoverable ram drive (RAD) back but I don't know if we will ever see another OS4 update.

@Snuffy

Quote:

Why? Because Snuffy said so in the topic you linked to?
I used the old RAD for years on my A4000 and on my ?A1/OS4 until OS4 final was released. It still works on my OS4 system but no longer survives a reboot. It was far more flexible than RESRAD and I would like to see it returned to it's original functionality.

On a classic Amiga, you can edit the RAD mountfile and create a RAD drive that you can "DiskCopy" floppy disks to for faster operation and manipulation. You can also mount multiple RAD drives if you want. You can copy your boot partition or boot floppy to a RAD drive and reboot from the RAD drive for increased speed. Unlike RESRAD, the size of a RAD drive was only limited by installed memory. I used large (48MB, 64MB) RAD drives on my classic Amiga and my ?A1 until OS4 Final appeared. RESRAD is limited to 15-20MB and slows down the cold boot process. I'm using a ?A1 but I'll bet that you can still diskcopy floppies to RAD on a classic Amiga with OS4 installed.

@Gilloo

I made a small OS4 makefile and compiled your sources for OS4 and tested the resulting programs. They seem to work but GCC issues warnings for almost all the printf statements. I don't want to post them all here but here is an example:

usblist1.c:78: warning: format '%ld' expects type 'long int', but argument 2 has type 'int'

Almost all the arguments have type int but the printf statements use %ld or %lx. Since you are using a different compiler and includes for 68k, the OS4 includes may be different and use different type definitions. Possibly your compiler just doesn't generate warnings for these kinds of errors. If you intend these files to also be tutorials for OS4 then the warnings should probably be eliminated by using the correct argument types in the printf statements.

Here is the simple makefile I wrote to compile usblist and usblist1 with OS4 GCC:

CC = gcc
CFLAGS = -O2 -mcrt=newlib -Wall
LDLIBS = -lauto
DEFS = -D__USE_INLINE__

.PHONY: all clean

all: usblist usblist1

usblist:
$(CC) $(CFLAGS) $(DEFS) usblist.c unistr.c -o $@ $(LDLIBS)

usblist1:
$(CC) $(CFLAGS) $(DEFS) usblist1.c unistr.c -o $@ $(LDLIBS)

clean:
@delete >NIL: *>NIL: usblist usblist1 QUIET

Here is the output from usblist1 -v:

USB opened
USBSysBase at 0x6ff7fdd0
[1] rawifc at 0x6ff06510
product="Front Roothub (unit 1)"
DEVICE (id=1,len=18)
USBVer 0100
Class 9
SubClass 0
Protocol 0
MaxPacketSize0 64
VendorID 0000
Product 0000
DevVer 0100
ManufacturerStr 0
ProductStr 1
SerialStr 0
NumConfigs 1
CONFIGURATION (id=2,len=10)
TotalLength 27
NumInterfaces 1
ConfigID 1
ConfigStr 0
Attributes 64
MaxPower 0 (0 mA)
INTERFACE (id=4,len=9)
InterfaceID 0
AltSetting 0
NumEndPoints 1
Class 9
SubClass 0
Protocol 0
InterfaceStr 0
EP (id=5,len=8)
Address 81
Attributes 03
MaxPacketSize 64
Interval 255 ms
INTERFACE (id=4,len=9)
InterfaceID 0
AltSetting 0
NumEndPoints 1
Class 9
SubClass 0
Protocol 0
InterfaceStr 0
[2] rawifc at 0x6ff06310
product="Rear Roothub (unit 0)"
DEVICE (id=1,len=18)
USBVer 0100
Class 9
SubClass 0
Protocol 0
MaxPacketSize0 64
VendorID 0000
Product 0000
DevVer 0100
ManufacturerStr 0
ProductStr 1
SerialStr 0
NumConfigs 1
CONFIGURATION (id=2,len=10)
TotalLength 27
NumInterfaces 1
ConfigID 1
ConfigStr 0
Attributes 64
MaxPower 0 (0 mA)
INTERFACE (id=4,len=9)
InterfaceID 0
AltSetting 0
NumEndPoints 1
Class 9
SubClass 0
Protocol 0
InterfaceStr 0
EP (id=5,len=8)
Address 81
Attributes 03
MaxPacketSize 64
Interval 255 ms
INTERFACE (id=4,len=9)
InterfaceID 0
AltSetting 0
NumEndPoints 1
Class 9
SubClass 0
Protocol 0
InterfaceStr 0
[3] rawifc at 0x6ff06650
vendor="Kensington "
product="Kensington USB/PS2 Wheel Mouse"
DEVICE (id=1,len=18)
USBVer 0110
Class 0
SubClass 0
Protocol 0
MaxPacketSize0 8
VendorID 047d
Product 102d
DevVer 0100
ManufacturerStr 1
ProductStr 2
SerialStr 0
NumConfigs 1
CONFIGURATION (id=2,len=9)
TotalLength 34
NumInterfaces 1
ConfigID 1
ConfigStr 0
Attributes 160
MaxPower 50 (100 mA)
INTERFACE (id=4,len=9)
InterfaceID 0
AltSetting 0
NumEndPoints 1
Class 3
SubClass 1
Protocol 2
InterfaceStr 0
HID (id=33,len=9)
bcdHID 0110
bCountryCode 0
bNumDescriptorType 1
bXDescriptorType 34
wXDescriptorLength 52
EP (id=5,len=7)
Address 81
Attributes 03
MaxPacketSize 4
Interval 10 ms
INTERFACE (id=4,len=9)
InterfaceID 0
AltSetting 0
NumEndPoints 1
Class 3
SubClass 1
Protocol 2
InterfaceStr 0
Bye!

I hope this is what you were looking for