Badly rendered icons in 9.1

a) 1280x1024 - 32bit - single monitor
b) no
c) no
d) ATI Radeon 9200 - Driver version 6.14.10.6360
e) Windows XP Home SP2

Is their any news about this issue? I also filed bugreport #350001465375 for this problem.

Regards,
Martin

No news here, at least. All we (the forum) know is that four people have seen the issue and two of them have ATI 9200 cards + XP while we don't know what the other two have.

I never had this problem, but I can reproduce it :slight_smile::

When I switch display depth from 32-bit to 16-bit mode, the problem appears. When I switch back, the problem disappears again.

Windows XP SP2
Screen resolution: 1280x1024
(other resolutions tested: same problem when switching to 16-bit mode)
NVIDIA GeForce 4 Ti 4600
No Remote Desktop, VNC etc.

In addition to my post above:
When I click the Favorites-menu in Internet-Explorer (v6.0) I also notice some Icon-problems:
(Left picture: 32-bit, right picture: 16-bit)

[quote="Christiaan"]I never had this problem, but I can reproduce it :slight_smile::

When I switch display depth from 32-bit to 16-bit mode, the problem appears. When I switch back, the problem disappears again.[/quote]
The others are saying they see the problem all the time though, even in 32-bit mode. There are some problems with icons in Opus (and Windows in general) when you switch from 32 to 16-bit (e.g. using Remote Desktop) but I think these are two separate (possible related) issues. If you restart Opus after changing the display depth it should be fine.

Yes, you are right, restarting DOpus solves the problem...

Since the problem seems to relate to ATI 9200 cards I have found one on eBay and ordered it to try and help track down the problem. Jon from GPSoftware says he's been looking for one in Australia without luck so if I can't work anything out here then the worst case I'll post my card to him so he can try.

Thanks for your great commitment Leo.

Leo,

I don't want to spoil your ebay adventure but underneath is my configuration WITHOUT the supposed "culprit" ATI card and WITH the same problem. It's my work PC btw.

a) 1680x1050 - 32bit - single monitor
b) no
c) no
d) Intel 82865G (Onboard graphics) Driver: 6.14.10.4396
e) Windows XP Professional SP2

Example:

As you can see in the screenshots below (click thumbnails to enlarge), I cannot reproduce the problem on my XP Pro machine with ATI 9200 and 6.14.10.6614 drivers.

Both the favorites menu and the translucent hidden file icons draw fine. The two screenshots show this and also the version of the ATI drivers and the display settings.

The only way I could reproduce what the screenshots in this thread show was by switching the desktop to 16-bit colour, but restarting Opus after doing so made the problem go away.

I tried changing some toolbar display options in Opus but I couldn't find anything that broke things.

So if everyone seeing the problem is absolutely sure that they're running in 32-bit colour (as shown in Catalyst Control Center in my second screenshot) and not changing between 16-bit and 32-bit at all, then I'm at a loss.

It's worth noting that the drivers I downloaded seem to be newer than the ones people with the problem have reported. Not sure why that is when ATI's website says they haven't been updated since 2006, but maybe the website means that was the last major update and there have been minor ones since then? Either way it might be worth a try to install the latest drivers from ATI. (BTW, that link is only valid for XP and 9200 users.)

[quote="jvolkering"]WITHOUT the supposed "culprit" ATI card

...

d) Intel 82865G (Onboard graphics) Driver: 6.14.10.4396[/quote]
That version number looks suspiciously like an ATI driver version number.

The Intel drivers for the 82865G seem to be numbered differently, e.g. 14.17.0.4396 is the latest one:

downloadmirror.intel.com/9498/ENG/relnotes.htm

Found via the 82865G downloads page

Perhaps I am looking at the wrong version number and Intel and ATI's driver versions are coincidentally the same down to the 4th component?

actually Leo, most intel driver version numbers seem to have the 14.10 numbers cause the driver version on my laptop also has those numbers (Intel centrino vPro chipset).

NEW theory!:
if (version_number == sSecret_numbers)
{
DOpus.AlphaBlending == false
}

:stuck_out_tongue:

...maybe my boss put an ATI card in my worksystem without informing me. I could have been playing games @ work al this time!...well kind of.

happy hunting. If you want some more info. Let me know!

Grtz Jeroen

Yes, but see your post. You said your drivers were 6.14.10.xxxx not 14.10.xxxx.yyyy. Was that a typo?

jvolkering:

Fix your email address - all the topic reply notifications are bouncing back to me - they are being rejected as spam...

@ steve: I have to FIX my email address? WTF? It's a normal address I used for years and years. So the answer is "no can do"! Fix your spamfilter....

@ leo: No typo. This topic is kinda getting offtopic. I don't have a problem with the alpha issue but i'm trying to help you guys. Reading the reply by steve it isn't appreciated CHANGE YOUR EMAIL WTF...sorry, frustration... :slight_smile:

Last time i'm trying to help:
You think it has something with a ATI videocard. I don't have one in the system which has the problem. You're making an assumpion that it's something with the driver and are making a direct link with the version number which coincidentally is "similar" at best. I'm trying to tell you that it's has nothing to do with that by saying that on my laptop (which doesn't have the problem) the driver also has 14.10 in the version number. Version number on laptop is 7.14.10, driver version on work pc is 6.14.10. My point was that this is maybe an Intel numbering scheme which has absolutely nothing to with the problem. Just to help you. Don't want you to invest a lot of time on a wild goosechase.

But back to the problem. The problem started with DOpus 9.1. DOpus 9.1 has a big change in the way icons are drawn according to the releasenotes. I'm thinking Jon did do some drawing stuff himself instead of using prefab GDI functions to circumvent the problem with the contextmenu's (Just a guess, what do I know about it...). My guess it has something to do with that.

But I think I'll stop responding, because I'm irritating steve's (actually his emailclient) too much... yes,pun intented. Sorry steve :stuck_out_tongue:

Your email provider is seeing the notifications as spam - hence it's bouncing the notifications back to me. Were you not even slightly curious as to why you never received the notices?

Either fix it or change addresses - this is not open for negotiation...

(I've manually removed your subscriptions to the two topics)

For your info - here's the error I am receiving:

Final-Recipient: rfc822; jvolkering@XXXX.XX Action: failed Status: 5.0.0 Diagnostic-Code: X-Postfix; host mx.XXXX.XX[XXX.XX.XXX.X] said: 550-Spam refused: tanis@fud.dreamhost.com does not have a valid domain (no MX 550 record). (in reply to RCPT TO command)

(I've hidden your domain and the IP in the above error report)

jvolkering, I'm confused though. You said you experience the problem on a machine with:

But Intel's web page for that chipset has graphics drivers versioned 14.10.xxx.yyy, not 6.14.10.xxx.

So am I looking at the wrong webpage, or do the drivers report a different version number once installed, or did you type the wrong chipset and/or version number?

Obviously I'm not saying that only ATI use 6.14.10.xxx version numbers; but from the web page it looks like the Intel 82865G doesn't use 6.14.10.xxx version numbers, which is why I am confused. I just want to be sure whether or not this is being seen on non-ATI hardware.

(The ATI web page calls their drivers version 6.11 and says they were last updated in November 2006, both of which differ from what is reported once they are installed, so it's possible Intel's web page is simply misleading in the same way that the ATI web page is.)

Hi to everyone following this thread.

9.1.1.0 just came out and has some fixes for toolbar icon rendering and shadows (alpha channels). Neither myself nor GPSoftware could reproduce the exact problem that some of you were seeing but we did find & fix some related issues so give the new version a try and let us know if things are better now.

Details of the release are here.

The download is here.

For me it's fixed with V9.1.1.0.

Thanks,
Martin