Unofficial CyberGraphX v4 bug fixes

[kamelito]

Why not just use the Aros version ?
Kamelito

[itix]

I am in the process of buying a new (old) graphics card for my Amiga A2000. Guess I get to wade through all of this quite soon. So far, I only have v3, so I need to upgrade to version 4 where?

You can still continue using CGX 3. Version 4 is better, for example it has better async blit support, 56KHZAUDIO env var and loads of other things but updating is not necessary.

[guest11527]

That sounds very reasonable and all, I just don't see the world working in that way.  While we're at it, can we get Holger Kruse to post up publicly that he's abandoned Miami?  Stefan Stuntz to stop skiing for a moment to come to Amiga.org and post that he's abandoned MUI?  

Do you have reasons to work with or on their files, and re-publish the binaries? I believe Stefan is alive and well, so if you want to ask him a question on MUI, I doubt he'll refuse to answer if you ask politely. I know nothing about the whereabouts of Holger, but again, just being too lazy to ask does not mean that you get rights on his programs.  

It seems to me, and granted this is just my experience and all, but when a person abandons something, they don't come out and make some grand statement that we can all reference later.  Usually they just walk away.

Maybe they don't make statements when they go, but why does that stop you asking?  

I don't even use CGX.  I don't care and have no dog in this fight.  I found P96 works much better for my purposes.  But if I was a classic user running CGX and hadn't seen an update in 7+ years, I'd be glad as he** for someone like Cosmos to come along and release a bugfix version.

If you would be an author, and would find that people just patch around in your software without asking you, what would you do? And what would you do if people approach you nicely and ask first whether they may make modifications? Just put yourself into the position of a developer for a time who left the Amiga a while ago, then think which case appeals more to you. See?

[guest11527]

I doubt Frank Mariak would have much interest in repairing something he wrote that long ago.
Besides, even if he still has an Amiga, he already has his hands full developing for MorphOS.

Maybe so. But don't you think it would be Frank who should decide on what should happen with his work?

[guest11527]

I am in the process of buying a new (old) graphics card for my Amiga A2000. Guess I get to wade through all of this quite soon. So far, I only have v3, so I need to upgrade to version 4 where?

Depends on the graphics card you want to buy. For many graphics cards, we have two options: P96 and CGfx. The former is still "sort of" supported, at least inofficially so, and not by its original authors - but with permission of the original authors.

[kamelito]

It would be better use SetPatch() to patch this library vector to save D2 before calling original function and restore D2 on return. However, developers knowing said function clobbers D2 can work around it without resorting to tricks (IIRC the result is D2 = D2 + width).

Anyway, CGX is not abandonded, it is part of MorphOS, and the latest update was just few weeks ago. It is copyrighted stuff and Cosmos has no right to release "new" CGX4 versions.

Did you mean SetFunction()?
Kamelito

[Oldsmobile_Mike]

Do you have reasons to work with or on their files, and re-publish the binaries? I believe Stefan is alive and well, so if you want to ask him a question on MUI, I doubt he'll refuse to answer if you ask politely. I know nothing about the whereabouts of Holger, but again, just being too lazy to ask does not mean that you get rights on his programs.   Maybe they don't make statements when they go, but why does that stop you asking?

Aside from a little Commodore 64 BASIC and some VB in college, I'm not a coder, so no, I'd have no reason to work on or re-publish their binaries.  I was just using them as examples.

Stefan, for example, from multiple forum searches I've found many posts from people who've already tried to contact him.  Laugh at me all you want, but when I found this website this past May I went ahead and sent him the Paypal registration fee.  When I was a broke teenager I pirated the heck out of his software.  Now it just felt right to "make up for that", in a way, by registering.  I got no response at all, not even a "hey, thanks for the money".  Which is kind of a shame, but I'm still glad I sent it, in a "doing the right thing" kind of way.

Around the same time I registered the WarpDT datatypes and a few other Amiga software packages that I've used over the years.  Again, just trying to do the right thing.  Others did write me back, which kind of warmed my heart to know they appreciated the gesture.

Holger, on the other hand, I believe he "flipped the collective bird" to the entire Amiga community some years ago, and dropped off the map.  No longer accepting registrations or even willing to discuss his Miami software, assuming you could even track down his contact info.  Which is a damned shame, because I would gladly put money in his hands today to see that software continued to be developed.

But I guess for some people, pride is more important than money.  I don't know, just hypothesizing.

[Jose]

Just to be clear, I don't think distributing binaries as a new CGX version is right, because that is distributing it almost as if it's the official continuation of the original without the authors permission.
But I think a separate patch does no harm.
On the other hand I also think contacting the author first would be the best and more correct, although it's pretty evident what the likely outcomes are.

[LoadWB]

Stefan, for example, from multiple forum searches I've found many posts from people who've already tried to contact him.  Laugh at me all you want, but when I found this website this past May I went ahead and sent him the Paypal registration fee.  When I was a broke teenager I pirated the heck out of his software.  Now it just felt right to "make up for that", in a way, by registering.  I got no response at all, not even a "hey, thanks for the money".  Which is kind of a shame, but I'm still glad I sent it, in a "doing the right thing" kind of way.

Around the same time I registered the WarpDT datatypes and a few other Amiga software packages that I've used over the years.  Again, just trying to do the right thing.  Others did write me back, which kind of warmed my heart to know they appreciated the gesture.

Holger, on the other hand, I believe he "flipped the collective bird" to the entire Amiga community some years ago, and dropped off the map.  No longer accepting registrations or even willing to discuss his Miami software, assuming you could even track down his contact info.  Which is a damned shame, because I would gladly put money in his hands today to see that software continued to be developed.

I am in the same boat for the most part.  Although I was actually able at some point to register MUI on one computer, then I installed it on several.  Many years back I registered for a few copies of MUI but never received a key, so I use the same key on multiple machines with the knowledge that I have done right.

And I unfortunately feel the same way about Holger.  When I was broke I actually spent the money on Miami (as well as YAM, AmIRC, and AmFTP at the time -- saved up to do them all.)  I think it was a bogus update to WildIRC (the ARexx script bundle) that stole my Miami key and replaced it with a generic "Digital Corruption" key.  When I contacted Holger about it, he berated me for pirating Miami (even though I still had the email confirmation of my registration!)  Then once he went over to Rebol he pretty much gave up on all of us.  That said, I actually have not had to "pirate" Miami as between Genesis, a registered Miami Deluxe I purchased, and now Roadshow, I have all of my machines well covered for TCP/IP.

But I guess for some people, pride is more important than money.  I don't know, just hypothesizing.

It's true, even stubborn pride.

[Cosmos Amiga]

Maybe it has been abandoned. However, wouldn't it be much nicer to hear this statement from the author that (apparently) abandoned the project? And then post that answer here? At least that would help everybody else to make a decision which product to pick (P96 is not quite that abandoned, for example) or how to handle the software.

So for example, if we hear from the authors "year, do whatever you want with it", then at least this could possibly motivate another group to pick up the project and start working from it.

Just guessing... well, I don't know. It seems *likely* that you are correct, but I believe there's only one person that should decide. Hint: None of us here.

In my .readme, I give the name of two d2 bugged functions... So if the author really want to fix, he can by himself very quickly : this job need few minutes only...

As I explained, there is a conspiracy to destroy the Classics by all the possible ways. Classics are too good computers, some don't like that... It's my point of view, if you are disagree, you are a naive man for me...


About Warp3D, there is a bug in the 3dfx GPU Avenger, some internal deadlock/crash when using some registers in a certain order... So, the GPU do an hard reset, and after that, can continue... Bug fixed in the VSA-100...

I see this bug with the Prometheus driver and my optimized W3D 3dfx libraries : some clipping appear... No clipping on my two Mediator configurations, maybe Elbox bypassed this bug, I don't know exactly...  

I sent two emails to Bigfoot (a member of the MorphOS team, like Frank Mariak ?) who have found and understand this bug in the MOS 3dfx drivers for having more informations : no answer...

[stefcep2]

No, lousy work. Really, guys. There is no reason to applaud. There's a reason to be mad about it. It's another sad example how the "community" (or absence thereof) threads its developers (or ex-developers).

It would have been just so easy: Simply ask the author (Frank?) about the bug, ask him kindly to fix it, or provide ideas what to do about it, or for permission to fix it. Nothing is easier than that!

Consider you've created a painting or a building (as an architect) and somebody else comes along and repaints the eyes or puts another roof on top. That's simply not how you treat people that invested quite some time and had a hard job completing the building or painting in first place. You just damn ask them, that's the absolute minimum I would expect.

As an architect, you are even protected by law from third-party modifications on your creation, I believe it's not asking for too much to show the same level of respect to software architects.

If the answer of the architect would be "go, p*ss off!", you can still react on that, but in most cases, it was not when I tried (with exceptions, of course). And, on the plus side, it keeps such developers motivated to invest time in their work.  

Stuff like that - pirating other's people's work and messing with it in the way they may or may not have intended - is not going to help! Actually, it may seem to help on a short time scale, but it will motivate nobody to invest some time into their old projects. Bug reports, on the other hand, or hints for improvement, may! Yes, it takes longer, but yes, it may be worth at least giving it a try.  

I'm not even commenting on the correctness of the patch. I don't know. It's not my work, and not my bug. But the form of communication, and the way that some people even applaud on this is just something that makes *me* mad.

Folks, if you want to be part of a serious community, act professional!

Well if the architects creation resulted in a leaky roof and people living in that "creation" were getting wet, and some roof plumber fixes the leak for *free* because the architect wants nothing to do with his decades-old creation, what has the architect got to complain about?

Seriously. there is an important distinction between fixing damaged goods when the creator has no intention of doing it, and stealing the creators ideas, passing them off as your own and taking credit for the original work.

The latter is not happening, so I don't see what case the original author can make other than: "Its MY work".  

And if that is all what it amounts to, then as CGX 4 was a commercial product that I for one paid for, that gives me as the paying customer the right to say "Sure since it is YOUR work, then fix YOUR bloody work, or get out of the way and let someone else do it!"  

If coders want to be treated as professionals then part of the definition of a professional is that the buck stops with them for their work, and it is up to them to make good when their work is defective. If they don't want to be held to that standard, then they can't complain when someone fixes their work for nothing on the grounds that they didn't give permission.

[stefcep2]

I am in the same boat for the most part.  Although I was actually able at some point to register MUI on one computer, then I installed it on several.  Many years back I registered for a few copies of MUI but never received a key, so I use the same key on multiple machines with the knowledge that I have done right.

And I unfortunately feel the same way about Holger.  When I was broke I actually spent the money on Miami (as well as YAM, AmIRC, and AmFTP at the time -- saved up to do them all.)  I think it was a bogus update to WildIRC (the ARexx script bundle) that stole my Miami key and replaced it with a generic "Digital Corruption" key.  When I contacted Holger about it, he berated me for pirating Miami (even though I still had the email confirmation of my registration!)  Then once he went over to Rebol he pretty much gave up on all of us.  That said, I actually have not had to "pirate" Miami as between Genesis, a registered Miami Deluxe I purchased, and now Roadshow, I have all of my machines well covered for TCP/IP.



It's true, even stubborn pride.

Its things like this that result in no sympathy from for developers.  If they want to be treated as professional, don't release bugged software.  If you do, then fix it.  If the bugs are found many years later, tough for the developer, je has an obligation to fix it regardless.  I don't see Takata trying to to get out of replacing faulty airbags 10 years or more after the bags were installed.  If its faulty on day one, its still faulty 10 years later, how long it took to find the fault is irrelevant, its their job to fix it.

But that's not what happens- "oh, support has ended, bad luck".  Well then don't cry when others reverse engineer your product to fix your defects.

[Trev]

I don't need the architect's permission to add on to my house. I don't need the hardware designer's permission to add on to my Amigas. I don't need the software designer's permission to add on to my software, particularly when the operating system was designed specifically with that capability in mind.

There are many friendly, open, and reapsonsive developers in the community. There are also those who operate as a professional cartel, stonewalling attempts to effect change. It's sad, it's counterproductive, and it's demoralizing to those who donate time and resources to the betterment of the community without any desire for compensation.

[spirantho]

I appreciate Cosmos' intentions in this, but quite apart from the politeness of asking the author, there are more practical reasons for this being a bad idea.
It's simply this: If Cosmos releases a v4.8 of a library as an enhancement to v4.7, and then the original author suddenly reveals he has his own new version v4.8, then all of a sudden we have two v4.8 versions, which are different, by different people, and have different capabilities.
And then "hilarity" ensues. And by "hilarity" I mean a complete mess.

Do the update via a patch, not a replacement. Then if the author takes it upon himself to fix it, the patch can just be removed.... but having two different libraries - or even the possibility of two different libraries - with the same name and version number, but with different APIs... that's just asking for trouble.

[guest11527]

In my .readme, I give the name of two d2 bugged functions... So if the author really want to fix, he can by himself very quickly : this job need few minutes only...

As I explained, there is a conspiracy to destroy the Classics by all the possible ways. Classics are too good computers, some don't like that... It's my point of view, if you are disagree, you are a naive man for me...

Cosmos, you're one of the people that destroy the classics, don't you see this? You're destroying the motivation of investing any time into software simply by taking other's people's code and ripping it apart. That's not how development works. It creates trash, unmanageable versions, unmanageable software, it defeats any sane development process.

A readme is nice, but what on earth is so hard and use the very same readme and send it to the author fist *before* you proceed with publication. That's done in five minutes. Who stops you doing that? It is just a matter of politeness and respect, leave alone we would then know what the author thinks about it and not what you *believe* he thinks about it.

No, I'm pretty sure that there is no "conspiracy" except in your head. People leave because they lost interest, probably because there was no longer a motivating factor. Probably motivation was gone exactly by activities like this.

[Brian]

If the classic version is abandoned by the authors, the authors doesn't even bother to reply when given all information in order to fix the issue then I very well understand why Cosmos does this... it's because he can. He could just as well sit on the fixes and enjoy them himself but he choose to share them and if a users decide to use his work then apart from enjoying it they should understand that there might be compability issues now or down the road. No more no less.

I would like to see this distributed as a patch for the original file though.