Posts by supernobby

Hi Jens, thanks for checking. I also have OS3.2 already, but for some testing purpose would like to keep a plain OS3.1.

If possible, at least the installer and/or documentation could contain a note when installing on OS3.1 to get the picture.datatype from the P96 2.0 in Aminet to avoid trouble.

I found an older 2.4 version of IBrowse, but same issue.

Then I installed the P96 2.0 from Aminet and noticed, this comes with an updated picture.datatype v43. I replaced the original V40 picture.datatype with it and then IBrowse also starts on a P96 Workbench screen. P96 V3 does not include a picture.datatype update?!? Or did I miss something?

No, not two GFX cards in the system at the same time. I had the Cybervision64 installed and the problem happened. Now I have changed to a ZZ9000 an still can reproduce the issue. So it could be something with P96.

I have an AMIGA 4000D with 3.1 ROMs, Cyberstorm MKII and a ZZ9000 or Cybervision64. IBrowse is the latest version 2.5 and P96 is 3.3.1. IBrowse opens on the Workbench screen.

Problem is, the IBrowse is crashing/freezing the whole system when launching and the Workbench screen is a P96 screen. So far I can say, this happens when it loads ilbm images. If I remove all ilbm images from the program folder, IBrowse complains about missing images, but starts and can be used.

When Workbench is a native AGA screen, IBrowse starts normal also with the ilbm images in place.

When softkicking OS3.2 an using OS3.2, IBrowse also starts normal with ilbm images in place when Workbench is an P96 screen.

So, something seems wrong with the combination P96+OS3.1+IBrowse2.5. Could P96 be the problem? If I can help to find out more, please let me know.

And if somebody else has this configuration at hand it would be great to know it it works for others or as well not. Thank you.

Hello,

I have recorded the USB traffic between my rndis class and a phone with an USB analyzer. First using Deneb and then with RapidRoad.

Communication with RapidRoard "stopps", after a CDC OUT packet with 400 bytes. After that, no more CDC IN packets are received anymore. The recording shows some probably corrupted CDC OUT transfers instead. All this works with Deneb. There the CDC IN transfers are recorded as expected.

If you want, I can send the logs and also the Poseidon class for testing this yourself.

Bye,

Andreas

Hi Jens,

I wrote this drivers myself, because there were no such driver in the package. They work with the same Poseidon software from the Wiki and the Deneb card, but not with the RapidRoad card. So, I would suspect some problem with the RapidRoad driver then if you can't confirm any differences from the programming API. Can't you help a bit with figuring this out? I will do my best to help as much as possible. Does the RapidRoad driver have some debug options?

Bye,

Andreas

Hello,

for some time I got a RapidRoad module for my XSurf100 in my A4000. Until then, I had (and still have) a Deneb in there and wrote 2 drivers that used Poseidon. One does an App Binding (a VHI driver for an old webcam using isochronous transfers) and the other is a normal Poseidon class (network driver for rndis devices). Both work when connecting the devices to the Deneb. But, when trying both on the RapidRoad, unfortunately, both drivers do not work. That is of course a bit disappointing. The Poseidon Software is 4.5 and the RapidRoad device is 1.7, which should be the current version.

Before looking deeper into this, I want to ask, if there are any differences for programmes, that I should be aware and that could explain, why my stuff does not work with RapidRoad, but work with the Deneb. If there are none, what would be the best approach to identify the problem, which could be on either side.

The RapidRoad generally works, as I can use USB stick or mouse.

Thank you!

Andreas