[Events] Elena Novaretti (ZoneXplorer - PowerIcons) is present with a Peggy II at WEBBIT | ANN.lu |
Posted on 08-May-2004 16:00 GMT by Raffaele | 49 comments View flat View list |
I have news that Elena Novaretti, author of good software programs such as ZoneXplorer and Power Icons (and a known Pegasos fan and owner) is also present at Webbit Padova with a Pegasos II kindly furnished by Pegasos Italia just for the exhibition event.
We expect soon impressions of Webbit show from Elena, and news about the impact which Peggy II made to italian public.
Wow!!!
After 10 years of absence to important events here in Italy, two Amiga based platforms are there!
More! I have news that the booth by Virtual Works, Ikir Sector, Soft3, Bitplane (with partecipation by Eyetech, ZetaOS and Cloanto people) IS A DOUBLE-SIZE booth.
Very impressive.
And more, seems that (basing upon Ikir statements on the web), there are two Pegasos at Amiga booth...
Amiga has a Very Very strong presence and a lots of opportunities to impact the public at Webbit Italy.
Good news for our beloved platform.
|
|
List of all comments to this article |
Elena Novaretti (ZoneXplorer - PowerIcons) is present with a Peggy II at WEBBIT : Comment 36 of 49 | ANN.lu |
Posted by Sigbjørn Skjæret on 09-May-2004 15:27 GMT | In reply to Comment 35 (Kjetil): "forcing a bad program form trashing memory is not a crash, it's a program termination."
To the user that is equivalent to a crash.
"68k assembler code, it will be translated by jit to PPC, the jit compiler will only compile it once, there is no need to translate it the 2en time the code is executed, when you start the program you might get an hi overhead, once it starts running it will be come more and more PPC native, so the overhead becomes less,"
So, it doesn't translate all at once (since that would make programs start really slowly), right? That means while this code is not translated yet there will be alot of exceptions generated by jumping in and out of OS functions that are PPC... Also, what about 68k libraries? Last I heard Petunia doesn't know how to JIT those (if someone knows the current status of this, please fill us in) .. now that will certainly generate alot of exceptions .. and hey, what about those nice "traps" (jmp.w 0, wtf is that about?) that are added in the 68k libentries for PPC libraries, those must make a nice overhead (it parses 68k registers for dawgs sake!) too. :P
Now, as an illustration of the (virtual non-)impact an extra api layer can have, you just have to look at the Warp3D wrapper in MorphOS (which Ben Hermans described as "A replacement based on a stone-age old technology called Rave which was ditched by Apple years ago." (that should give you some perspective from your angle)) .. on the same hardware it not only is fast, it's actually many times faster than the original! Why? Simply because the layer is efficiently written, and it's running on top of an api that is cleaner and more efficient than the layer api (ofcourse, better drivers help too). As you can see, if done right, layers can have next-to-no impact on performance, and whatever insignificant loss there is can quickly be recouped by having a clean and efficient base...
Anyway, the point is that you are (once again) discussing things of which you know nothing about (completely offtopic to the thread even (ok, you didn't start the offtopic discussions, but you did propagate them by adding comments contradicting all other evidence)) .. you admit that you have no first-hand (or even second-hand) knowledge of how MorphOS works, still you are always there to tell everyone how inferior this and that in it is, and now even going so far as to showing your blatant ignorance on how an MMU works (no, the overhead is not limited to when you get exceptions), yet talking as if you were some kind of authority on the matter...
I've kindly suggested to you before that perhaps you should try to read up and make actual observations on matters before you comment on them before, but that obviously failed (and you seemed to have taken some kind of offense to it as well, judging by some other threads), now I will kindly urge you to do so for your own good, however I suspect it won't help. :(
- CISC |
|
List of all comments to this article (continued) |
|
- User Menu
-
- About ANN archives
- The ANN archives is powered by #AmigaZeux. It was updated daily (news last: 22-Oct-2004; comments last: 18-May-2005).
ANN.lu was created, previously owned and maintained by Christian Kemp, www.ckemp.com.
- Contribute
- Not possible at this time!
- Search ANN archives
- Advanced search
- Hosting
- ANN.lu was hosted by Dreamhost. Sign up through this link, mention "ckemp" as referrer and he will get a 10% commission on any account you purchase.
Please show your appreciation for any past, present and future work on ANN.lu by making a contribution via PayPal.
|