I would say that quick'n'dirty ports feel alien (like running X11 apps through cygwin on Windows), are more unstable that native ports, feel unfinished and integrate badly with the host OS (like Cygnix X11 stuff)
That's not a definition of a quick'n'dirty port. A quick and dirty port refers to a port done in a limited time, without a good quality check of the port and leaving bugs mostly untouched. You cannot judge poor an X11 port, since by its nature it will feel alien compared to its host window manager. In this case the port has a good integration with Workbench within its limits, just check the work that has been done to get the gadget graphics as similar as possible with OS4's one. Regarding unstability I can only speak of my own experience, at least AbiWord and Gnumeric run pretty stable on my system. And still, being a port you can't really judge on stability, unless you pretend that a port in order to be a good one must have fixed all the stability bugs of the original code. And for the last part, what part of these ports make you feel them unfinished?
Do you really call "useable" and "polished" having to use windows keyboard shortcuts to select menu options because GUI doesn't respond to mouse events...
We talking here about how "useful" it can be, not about "useability" and "polishness". And yes, having to use keyboard to select menus doesn't make it less useful, since the plugin I'm using does the job well.
You should realise that most PPC software is just a collection of SDL/CLI unix ports that usually run better in posix platforms. In case port is not easy to do it's common practice to take MorphOS code: MPlayer, BasiliskII, Freespace2 and others...
I'm still interested to know how have you found the information that all the SDL ports have been made by just typing "make". The part that they usually run better on other platform is IMHO not true: it is true for those programs that require more CPU power than the available one on current OS4 machines. But at the moment I can't recall a single SDL application that runs sluggish on my system. There have been some ports that ran badly under OS4, but those ports AFAIK haven't been released. For the use of MOS code, I wouldn't call 3 uses out of some hundreds ports "common practice".
It's normal, just like if you ask me if Dunlop tires are good and I reply you that I prefer pirelli ones
No, for that I would have asked you "how are Dunlop tires compared to other ones", while in this case it's a case of "I want to try Dunlop tires, how are they?" OP has even explicitely stated that he's not interested in a MOS comparison.
Since Efikas are being sold new with its 2 years of warranty and Sam440 has just been produced in small batches and is not produced continuously (I doubt they produce any now that Sam460 is planned) they have the same status now: sold new with warranty but not produced actively.
Efikas will never be again produced. Sam440 are still produced (and that has been confirmed by ACube, even if the 460 will enter production). I see a difference here.
Price increase can not be justified because a machine produced for the same small market 5 years ago costed the same and it's named Pegasos2/G4. Both run OS4 and The old 5 years-old machine is faster. Quite sad if you ask me.
Have you thought about how's the PowerPC CPUs market today in respect to 5 years ago, and have you taken in account that the major CPU buyer that existed 5 years ago isn't anymore a buyer these days?
Sam's advantages are shaded by OS4 (lack of) features: USB works slower than Efika and 3D is probably slower too despiting higher clockrate. It has more ram and faster HD access but I'm not sure I want to pay 450Euros to get a system marginally better.
I guess that everyone have its own opinion whetever that is marginal or not.
Sam G2 cpu technology is obsolete and not many customers would expect having to pay 500Euros for it.
They are enough to justify the release of a new platform, though.
But MorphOS does and it runs classic software probably better than 533-600Mhz Sams.
But OP isn't interested on MOS.
I'm implying most of people paid large sums of money for A1 and specially Sam440 because it ran OS4, otherwise the hardware would have been rejected and they would have bought better hardware like Pegasos2 or Mac Mini.
They had their right to choose. At the time they had the choice to not buy A1 and OS4 if they felt that the hardware was too pricey.
hackish means using an embedded cpu and selling it in a desktop motherboard with limitations you would never expect in a motherbaord like having just ONE SATA that gets disabled if you fit a PCI-e card.
And I'll repeat my question: are PCI Express gfx cards incompatible with 4x lanes slots, and work only in 1x ones, to make you force to plug it in Sam460's 1x lanes slot?
All A1s are Teron prototypes hardly tested. Sams hardware looks better although it's obvious that it has not been tested as deeply as big companies like Apple test its hardware.
What has ACube to do with MAI? Regarding the tests, the question is if they have done enough tests for a releseable product, which neither you nor me can know. So you can't really say that SAMs are "prototypes" and "hardly tested".
Other betatesters and users who sold their (u)A1s to buy Peg2/G4 instead of Sam440 may not agree with you.
That still doesn't make your opinion as a general rule.
I guess it's the same people who's happy using alien x11 ports on their "Amigas"
And what's wrong with that?
Varthall
