29-Mar-2024 15:05 GMT.
UNDER CONSTRUCTION
Anonymous, there are 427 items in your selection (but only 327 shown due to limitation) [1 - 50] [51 - 100] [101 - 150] [151 - 200] [201 - 250] [251 - 300] [301 - 350] [351 - 400] [401 - 427]
[Events] Notes about my presentation at the Alchimie 4 showANN.lu
Posted on 29-Sep-2004 10:40 GMT by Stéphane Guillard427 comments
View flat
View list
The html pages I made available was only a starter slideshow for my presentation. Here are a few words about what I shown, and what I said. Hi Gentlemen,

The html pages which are available at my site are just an introduction slideshow that I presented as a starter, it does not relate what I shown and said afterwards. Please find below a couple of notes about my presentation at Alchimie show.

Here is a quick summary of the presentation I did :

- I booted with boot delays set to 1 second for UBoot and for SLB (second level booter), thus less than 20 seconds after power on, we were with a fully loaded Workbench with Amidock.

- I started with the small html slideshow, presented on IBrowse. You can find those pages at the URL above. IBrowse loaded in 2 seconds with its About: page fully displayed. Browsing through the pages of the slideshow was instantaneous.

- While we were at it, I browsed the OS4 install guide, also as fast as can be, must say I also find this responsiveness impressive myself :)

Then I demoed as many things as I had time to during the 2 hours I had. Everything worked, fast & stable, and was smooth and impressed. I showed mainly:

- All os4 system, tools, utilities, prefs & stuff

- MUI, IBrowse, Amitradecenter

- Yam, SimpleMail

- ClassAction (M. Elsner's file manager)

- MakeCD

- MooVid running a DivX

- DVPlayer running an mpeg2

- AudioEvolution 4 with the demo project, cursors auto moving smoothly, the playback was smooth also, with mostly no CPU usage.

- AmiPDF with the AE4 manual and another heavy PDF file, very fast

- USB. I plugged a Microsoft IntelliMouse Optical into my USB hub, and we had 2 mice to control the system

- Crisot's slach5 winning demo (got applauses which it deserved)

- chip's rayrace realtime raytracer demo. When the juggler appeared, audience was impressed, but really took measure of what they were watching when I moved the mouse. Wait for the Altivec version !

- FPSE, AmiDog's PS1 emulator, with an oldskool game which ran nicely ('Cotton')

- some other visual toys I had at hand

The demos only grimreaped twice, and I anticipated the grims before they popped up :

- One was native glsokoban / w3d, when I launch it does a base page access (a “null pointer” bug in glsokoban). I didactically shown the disassembly which is available in the grimreaper window, was a store to r4, r4 was null etc. I clicked on continue, and it all went fine & fast.

- One was frying pan 0.3.1, I shown the app, and at one point I said 'now it should grim’ and it did. It still loaded fine though. I quit the app, clicked on reboot and less than 4 secs after, wb was up with amidock. That was the only reboot of the show.

I forgot to show (because of short time):

- Petunia... Almos, sorry, I had prepared something for that (side by side windowed jit & nonjit runs of voxelspace), but i both forgot, and was asked to stop at this point by the party organizers cause it was already 5:30 pm while I was scheduled until 4pm.

- ArtEffect

- USB with MassStorage (ie USB key or digital camera)

At the end, I had many fair questions to which I answered; my feeling is that the audience really appreciated the effort behind what I shown, and was conscious that we are not far away from a releasable 4.0.

Then came the expected question, 'and why doesn’t DMA work ?'

I said 'All what you saw was DMA since the 1st boot'.

I copied a few 100 MB files in a snap, with zero CPU (thanks to Pete Gordon for the clock/CPU docky, helps a lot). Then I switched to PIO, they saw 4 x slower and 80% CPU.

The audience understood that it was indeed DMA, and that was fast, and that was part of the overall smoothness of what I shown.

Then I explained the things below (this is the reference for my statements, please don’t rephrase or extrapolate or invent or whatever):

- IDE UDMA works on VIA and Articia on AmigaOne SE / XE / µA1 MK2 (as I demoed) ...

- ... except when the Ethernet chip goes online and is used.

- the Ethernet chip only triggers the problem, but it is not at all related to it (a test using a PCI Ethernet shows the same behaviour)

- We have made a driver for a Silicon Image 680 PCI IDE UDMA133 controller chip, this does UDMA 133 nicely, including when Ethernet is used at full speed.

- The fact that a PCI IDE controller solution works, shows that the problem is *not* related to Articia, since PCI DMA is *also* handled by the Articia, and that works.

- The full Alchimie show demo was done using UDMA, both from the VIA and from the Si680, without problem (but with Ethernet off, would the Ethernet have been turned online, I would have had to revert the VIA into PIO before).

- Things are currently under more investigation

In the meantime there are 2 options for existing A1 board owners:

- Use the VIA IDE controller in PIO mode when using Ethernet, and UDMA at other times,

- Purchase a faster (UDMA133) Silicon Image 0680 IDE PCI card (from around $20). This is my personally recommended option as the delivered speed is noticeable faster than the on-board VIA controller in UDMA mode.

Notes about my presentation at the Alchimie 4 show : Comment 101 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 13:41 GMT
In reply to Comment 99 (Christophe Decanini):
It can get difficult when your main criterion for whether someone is
wrong or right is based on who they are, and not on what they say.

(I'm not saying all OS4 supporters are like that of course, but some
are)
Notes about my presentation at the Alchimie 4 show : Comment 102 of 427ANN.lu
Posted by Thomas Frieden on 29-Sep-2004 13:43 GMT
In reply to Comment 77 (Johan Rönnblom):
> So in other words, OS4 is not really Amiga-compatible in this
> respect?

Well, if you want to put it that troll-ish: Yes.

The original exec had three entry points that where not public: Switch, Launch and Dispatch. These made up the scheduler. A task switch would run through all three of these functions:
- Switch, to retire the currently running task
- Dispatch, to ready a new task that will be running next
- Launch, to actually start the task again.

Some programs, like Executive for example, patch these functions with SetFunction. Therefore, they can alter the scheduling policy. The scheduler interrupt (or rather any interrupt) would end up running through the patched code.

Now scheduling is of course a time sensitive matter. Any schedule operation is dead time, the longer it takes, the more time is actually wasted. Therefore, this code is of course fully native, and should not be encumbered by any 68k code. And of course, this is why OS4 does _NOT_ run through the 68k entry points.

The effect is that it's no longer possible for a 68k program to take over the scheduler. Which is, IMHO, a good idea, since 68k programs would noly schedule 68k registers, and would of course be much slower than native PPC code.

Of course, this is an incompatibility. However, it's a very small sacrifice. How many programs hook into the scheduler ? Executive... PM. Oh yeah, Directory Opus has a CPU meter as wel.. I think the loss of compatibility is negligible.

So, either you knew that already and where just trolling, or you just made a fool out of yourself. Pick what suits you best.
Notes about my presentation at the Alchimie 4 show : Comment 103 of 427ANN.lu
Posted by Anonymous on 29-Sep-2004 13:45 GMT
In reply to Comment 95 (Carl Moppett):
> thats my understanding and what the screengrab shows sil0680 UDMA & onboard ethernet happy together.

Huh?
I didn't see any tests for data corruption there (like md5sum on repeated "real life" transfers of a large file), especially not while actually USING onboard ethernet (concurrent file transfer, plus a transfer rate benchmark).

On-board IDE with DMA is pretty fast too, it's just that you lose data, which you can't tell from a bytes/s number.

I'm not saying there IS a problem with "external" IDE as well, I'm just saying that the data people have brought forward including your screenshot is totally worthless to draw any conclusion from.

"Look here, it's fast" seems to be the standard reply to all queries about these issues. What's interesting is whether it's fast AND non-corrupting AND doesn't tax the CPU.
Notes about my presentation at the Alchimie 4 show : Comment 104 of 427ANN.lu
Posted by Christophe Decanini on 29-Sep-2004 13:50 GMT
In reply to Comment 100 (Stéphane Guillard):
If I was Hans-Jorg I could tell you that you are wrong because it may work with a VIA PCI IDE controller.I hope you got the point.

If someone say "sound is not working on A1" I would assume they are speaking about onboard sound. If they were speaking about sound or IDE udma in general I would assume that the board would be completely defective.

Instead of reading my post which in summary was "there is no point checking/testing/comparing the onboard driver when it is not working with ethernet" some people try to prove me wrong on rhetorics.
Notes about my presentation at the Alchimie 4 show : Comment 105 of 427ANN.lu
Posted by Rassilon on 29-Sep-2004 13:50 GMT
In reply to Comment 95 (Carl Moppett):
Carl,

you mentioned that the onboard UDMA ide and ethernet do not work correctly together. What exactly does this mean? Is it that writes through UDMA to HD get trashed, or is it more of a problem for data coming through the ethernet driver?

Cheers

Lewis
Notes about my presentation at the Alchimie 4 show : Comment 106 of 427ANN.lu
Posted by Thomas Frieden on 29-Sep-2004 13:51 GMT
In reply to Comment 88 (Fabio Alemagna):
Hello Fabio, nice to see you too. I was just waiting for you to pop up again.

> It's not really AmigaOS compatible in many respects, it seems.

So was OS 2.0 when it first came out, but of course, there wheren't so many trolls to herald that. Under 1.3, a lot of software didn't work anymore that worked under 1.2.

When moving from one revision to the next, some compatibility is always lost. It's a hard job to decide which compatibility is "worth" to be lost, and which should be kept. As much as I hate the term, it's about "moving forward". Loosing the ability for 68k tasks to measure the CPU idle time is a small loss, therefore it's been lost.

As such, your comment was completely pointless.
Notes about my presentation at the Alchimie 4 show : Comment 107 of 427ANN.lu
Posted by priest on 29-Sep-2004 13:53 GMT
In reply to Comment 100 (Stéphane Guillard):
While waiting for Eyetech response on how to fix this VIA UDMA - Ethernet collision ...

The si680 transfers 47Mb/sec (unoptimized UDMA drivers).
What is the speed of motherboard IDE with UDMA enabled? (with the same HDD, please)

((Jontsa told that with his HDD, he gets something like 6Mb/sec with PIO and >25Mb/sec with UDMA.))
Notes about my presentation at the Alchimie 4 show : Comment 108 of 427ANN.lu
Posted by Joël EHRET on 29-Sep-2004 13:57 GMT
In reply to Comment 76 (Thomas Frieden):
@ Thomas Frieden

have a look there http://www.pegasos.org... you see?? I suppose not,
and that's quite normal... but this website is running on a
pegasos1/G4 webserver running debian, and like it or not it has the
same Articia than current A1...

And bplan said in this place they ordered and received a small amount
of latest articias
Notes about my presentation at the Alchimie 4 show : Comment 109 of 427ANN.lu
Posted by priest on 29-Sep-2004 14:02 GMT
In reply to Comment 108 (Joël EHRET):
" And bplan said in this place they ordered and received a small amount
of latest articias"

I understood that they bought only "samples" of it.

After that they bought the remaining old ArtisiaS chips from MAI and put those to the final pegasos1 machines.

To me it sound stupid if they put those few sample chips on end user machines...
Notes about my presentation at the Alchimie 4 show : Comment 110 of 427ANN.lu
Posted by Christophe Decanini on 29-Sep-2004 14:03 GMT
In reply to Comment 108 (Joël EHRET):
"And bplan said in this place they ordered and received a small amount
of latest articias"

Which are the one used in A1 (and before someone say wrong let me specify in A1 sold with the new release of the articia S).
His point is that it is not exactely the same articia S as it is a newer release.Bplan's point is that they tested the newer articia and still found problems. Hence the Pegasos 2 with the Marvell.
Notes about my presentation at the Alchimie 4 show : Comment 111 of 427ANN.lu
Posted by priest on 29-Sep-2004 14:04 GMT
In reply to Comment 108 (Joël EHRET):
btw...

"pegasos1/G4 webserver"

What is the reason why new G4 cards do not work on peg1?
Notes about my presentation at the Alchimie 4 show : Comment 112 of 427ANN.lu
Posted by Fabio Alemagna on 29-Sep-2004 14:07 GMT
In reply to Comment 106 (Thomas Frieden):
> Hello Fabio, nice to see you too. I was just waiting for you to pop up again.

You may have noticed I "popped up" in the other thread too, in your defense.

As you see, I simply say what I think, or what are facts. If that irritates you, I sugges you take a rest from the computer and go out to breath some fresh air.

You're way too touchy, even if you try hard to hide it.

> > It's not really AmigaOS compatible in many respects, it seems.
>
> So was OS 2.0 when it first came out, but of course, there wheren't so many
> trolls to herald that. Under 1.3, a lot of software didn't work anymore that
> worked under 1.2.

Perhaps those SW actually used undocumented features, or simply relied on side effects of bugs which couldn't really be kept.

Care to mention which SW you're talking about, and which are the incompatibilities you're thinking of? You know, it helps talking about real stuff.

> When moving from one revision to the next, some compatibility is always lost.

Really? Is that a law or something? Let's say things differently - and properly - instead: "When moving from one revision to the next, some people think it's ok to lose some compatibility".

As you see, it's still a matter of opinions what's the right thing to do, it's certainly not a law or somesuch that states that incompatibilities must arise when jumping over revisions.

You may want to notice that I've refrained from commenting on whether the fact AOS4 breaks compatibility is a good or a bad thing, nonetheless you've jumped off your chair as if I had said anything negative about it, up to the point of indirectly (as cowardy is too strong, it seems) calling me a troll.

> It's a hard job to decide which compatibility is "worth" to be lost, and which
> should be kept. As much as I hate the term, it's about "moving forward".

I see. So, you're saying that, say, MOS has not and is not "moving forward"? Because, you see, you just said that to more forward you need to lose certain compatibilities, and MOS hasn't lost them, as far as I know. I know you were not talking about MOS, but I'm kind of tired of your habit of not talking about anything concrete, so please, answer - politely, if possible - to my question.

> Loosing the ability for 68k tasks to measure the CPU idle time is a small
> loss, therefore it's been lost.

It may be true, it may be not, I haven't commented on it and I won't. I've just stated a fact, although I couldn't help but notice the hirony of the fact that the "official" AmigaOS cares less about backward compatibility than its clones do.

> As such, your comment was completely pointless.

Of course, we all know that the only useful comments are the ones that support you, your product and your company ;-)
Notes about my presentation at the Alchimie 4 show : Comment 113 of 427ANN.lu
Posted by Joël EHRET on 29-Sep-2004 14:08 GMT
In reply to Comment 111 (priest):
Current G4 cards are clocked at 133Mhz FSB, G4 cards for peg1 were
clocked at 100Mhz FSB...

the fact bplan had only samples (very few) explains why only a very
small amount of machines were built... and these were the mobo used
for the very few number of peg1 with G4 like the one GGS data has.
Notes about my presentation at the Alchimie 4 show : Comment 114 of 427ANN.lu
Posted by Thomas Frieden on 29-Sep-2004 14:09 GMT
In reply to Comment 108 (Joël EHRET):
> have a look there http://www.pegasos.org... you see??

Now THAT is proof enough for me. Of course, it's absolutely clear from the web site that it's running on the very same Articia S model that the Amiga One has. Of course, how could I not have seen.

Have you seen http://www.microsoft.com ? They run on the very same ASUS mainboard that a friend of mine has in his PC.

> And bplan said in this place they ordered and received a small amount
> of latest articias

And where does it say it's the same revision as used in the A1 ?
Notes about my presentation at the Alchimie 4 show : Comment 115 of 427ANN.lu
Posted by priest on 29-Sep-2004 14:10 GMT
In reply to Comment 113 (Joël EHRET):
Then. Any idea why the FSB clock is locked on pegasos machines?
Notes about my presentation at the Alchimie 4 show : Comment 116 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 14:11 GMT
In reply to Comment 102 (Thomas Frieden):
Thomas Frieden: Uh, scsispeed measures the CPU time by busy-looping at
a low priority. I don't really understand what you're trying to say
here. :-)

Anyway, it works under MorphOS. Is that an important compatibility
advantage? I don't know, but in this particular case it's interesting
to know the available CPU time, and we can't know that for the OS4
system as it isn't compatible with the standard Amiga
benchmarking program for measuring disk DMA speed and CPU usage.

I don't see how I would be "trolling" by pointing this out. I assure
you that I *am* genuinely interested in this measurement, and it's
pretty obvious that the lack of Amiga compatibility of OS4 is quite
relevant to that, don't you say?
Notes about my presentation at the Alchimie 4 show : Comment 117 of 427ANN.lu
Posted by Christophe Decanini on 29-Sep-2004 14:12 GMT
In reply to Comment 81 (Anonymous):
a) "There's no OpenBSD port for the PegasosII" b) "so BSD on the PegasosII doesn't work...."

If a) is right I do think b) is right.
Tell this sentence to a few people and ask them if you are wrong.
Notes about my presentation at the Alchimie 4 show : Comment 118 of 427ANN.lu
Posted by Joël EHRET on 29-Sep-2004 14:16 GMT
In reply to Comment 114 (Thomas Frieden):
because I saw the mobo in my own eyes (and the articia chip, and I
saw a current A1 no later than 3 days ago)... and you were able to do
the same if only you were a bit more open minded... this mobo was
shown during BAPS last year in rotterdam...

/me just remember that hyperion were the only exhibitor to bring back
their material during saturday to sunday night....
Notes about my presentation at the Alchimie 4 show : Comment 119 of 427ANN.lu
Posted by priest on 29-Sep-2004 14:17 GMT
@Thomas Frieden & @Joël EHRET

Stop the debate.

Itix said "...pre-April Peg1 board is random data corrupt."

That Peg1 has different ArtisiaS than what current A1s have.

Even if the http://www.pegasos.org pegasos uses the same ArtisiaS than current A1s, it's IRRELEVANT because it has April2 and is therefore INCAPABLE of producing this or that data corruption.


(and this thread is about wholly NEW UDMA propblem, causing .... what? AOS4 deadlock?)
Notes about my presentation at the Alchimie 4 show : Comment 120 of 427ANN.lu
Posted by Stéphane Guillard on 29-Sep-2004 14:18 GMT
In reply to Comment 116 (Johan Rönnblom):
lets kill it once and for all.

The scsispeed used on my si680 grab is a quick & dirty OS4 port by myself : I just removed the 68k asm file, and recompiled untouched. This is the reason why you dont have used cpu%. Had you looked carefully at the grab, you would have noticed.

Additionnally, 68K cpu% would make little sense in an emulated context.

Finally, the stock 68k scsispeed from aminet works out of the box on my drivers, yelds more or less the same figures, and provides a 68k CPU% indication, for udma modes its usually 97% free.

I just used the os4 port because thats what I had at hand.

Facts.

Regards,
--
Stéphane
Notes about my presentation at the Alchimie 4 show : Comment 121 of 427ANN.lu
Posted by Anonymous on 29-Sep-2004 14:19 GMT
In reply to Comment 118 (Joël EHRET):
Neko said they had a few late Articia-S chips ( the same as the XE ) and threw them in the bin and none made it into consumer boards. If you search ANN hard enough you will find the quote.
Notes about my presentation at the Alchimie 4 show : Comment 122 of 427ANN.lu
Posted by priest on 29-Sep-2004 14:19 GMT
In reply to Comment 116 (Johan Rönnblom):
You enjoy that so much, don't you.
Notes about my presentation at the Alchimie 4 show : Comment 123 of 427ANN.lu
Posted by priest on 29-Sep-2004 14:24 GMT
In reply to Comment 120 (Stéphane Guillard):
Damn you, you spoiled his fun. ;-)
Notes about my presentation at the Alchimie 4 show : Comment 124 of 427ANN.lu
Posted by itix on 29-Sep-2004 14:31 GMT
In reply to Comment 63 (Hans-Joerg Frieden):
And? What is fixed in later Articia S revisions? Yes, as I see this VIA IDE vs eth problem is not Articia S problem but just a design flaw no-one noticed earlier. But nevertheless both Peg1 and A1 got broken on-board IDE and need HW fix to run flawlessly. Genesi got shit because of ArticiaS bugs, Eyetech is getting shit because of mobo bugs.
Notes about my presentation at the Alchimie 4 show : Comment 125 of 427ANN.lu
Posted by itix on 29-Sep-2004 14:33 GMT
In reply to Comment 62 (Chip):
Mike Bouma I assume? Yes, I know and I fail to see how A1 got any advantage over Peg1...
Notes about my presentation at the Alchimie 4 show : Comment 126 of 427ANN.lu
Posted by Thomas Frieden on 29-Sep-2004 14:33 GMT
In reply to Comment 112 (Fabio Alemagna):
> You may have noticed I "popped up" in the other thread too, in your defense.

Didn't notice... which thread do you mean ?

> Perhaps those SW actually used undocumented features, or simply relied on side
> effects of bugs which couldn't really be kept.

Perhaps ?

> Care to mention which SW you're talking about, and which are the
> incompatibilities you're thinking of?

Are you kidding ? It's well known that a lot of software broke between 1.3 and 2.0. I don't need to go into details.

> You know, it helps talking about real stuff.

LOL. Yes, that's true.

> Really? Is that a law or something?

Nonsense.

> Let's say things differently - and
> properly - instead: "When moving from one revision to the next, some people
> think it's ok to lose some compatibility".

Ok, if you see it that way. The fact remains that for some things, it's worth dropping compatibility, and that is the case here.

> You may want to notice that I've refrained from commenting on whether the
> fact AOS4 breaks compatibility is a good or a bad thing, nonetheless you've
> jumped off your chair

I didn't jump off my chair. You overrate yourself.

> as if I had said anything negative about it, up to the
> point of indirectly (as cowardy is too strong, it seems) calling me a troll.

I also didn't call you a troll. I called Johann a troll. I said "where not so many trolls". That's as much calling you a troll as you refraining from commenting on good or bad compatibility loss...

But if you want to, OK: You are a troll. Every time OS4 is mentioned, you jump off your chair.

See ? Better now ?

> I see. So, you're saying that, say, MOS has not and is not "moving forward"?

Here we go. That's your typical way of "arguing". Where did I say MOS ? Where did I say AROS ? Right. I didn't.

> I'm kind of tired of your habit of not talking about anything concrete, so
> please, answer - politely, if possible - to my question.

Why should I be polite to you when you insult me, calling me a coward ?

And it's really great to see you accusing _ME_ of not talking about anything concrete. You NEVER talk about anything concrete. You always gloss or skip over parts where you would have to be concrete. You don't answer questions asked to _YOU_ when you don't want to asnwer them, but keep on bringing in stuff that's completely irrelevant to the point, like you just did by bringing MOS into the picture. That's why I think you are a troll (See, I call you a troll again, I must be getting bold).

But to answer your question: When I said "moving forward", I wasn't even thinking about anything else but OS3->OS4. It's my opinion that moving forward involves dropping compatibility. For example, we're currently working on a nearly constant time memory allocation system. It will definitely be incompatible to some programs, but heck, reducing memory fragmentation and brining down allocation time is worth the trouble... Oh, of course in my opinion... No law involved...

> Of course, we all know that the only useful comments are the ones that
> support you, your product and your company ;-)

Useful ? I said pointless... meaning it was not to the point.

Regarding the pointlessness of anti-OS4 postings: It should go without saying that I'm bias ;-) However, I think I was never really unreasonable. I am picky, sure, but that comes from my troll alergy. If somebody makes a constructive criticism about OS4, I'm all ears. However, up to now I completely failed to hear any constructive cricitism, neither from you, nor from anyone else. When "blue" people post about OS4, it's mostly "Has OS4 a design flaw" and then they bring up overlooked references to functions that have never existed.

So yes, useful comments are those that support me, my company or my product. But not necessarily by having the same opinion as me, but possibly having another opinion, and bringing over the _POINT_ of that opinion. I have yet to witness that. And that's the fine line between criticism and trolling.

Currently, you are trolling (third time ;) and that's become really obvious with your last remark.

You talk about politeness: If you want to be treated politely, do so yourself.
Notes about my presentation at the Alchimie 4 show : Comment 127 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 14:34 GMT
In reply to Comment 117 (Christophe Decanini):
You're (potentially) wrong. :)

Both FreeBSD and NetBSD are supposedly being ported, I've no idea if
any of those work on the Peg 2 yet, but logically they may. His point
is that BSD != OpenBSD just as UDMA != onboard UDMA.
Notes about my presentation at the Alchimie 4 show : Comment 128 of 427ANN.lu
Posted by Joël EHRET on 29-Sep-2004 14:35 GMT
In reply to Comment 121 (Anonymous):
@ anonymous

I gave you facts with a way to prove what I said (ask GGS-data about
which articia revision this peg has)
Now beleive what you want...
Notes about my presentation at the Alchimie 4 show : Comment 129 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 14:36 GMT
In reply to Comment 119 (priest):
And as I remarked in comment 26, I made a very similar presentation
myself with a Peg 1, without any April, and I didn't have such
problems. You know, just because a problem exists, doesn't mean it
happens all the time or is easy to reproduce.
Notes about my presentation at the Alchimie 4 show : Comment 130 of 427ANN.lu
Posted by syrtran on 29-Sep-2004 14:36 GMT
In reply to Comment 117 (Christophe Decanini):
Christophe, your logic fails you.

The two parts of this sentence:
a. There's no OpenBSD for the PegasosII (true)
b. BSD doesn't work on the PegasosII (this is undetermined, I believe
it to be false)

If b were true, then it would force a to be true. However, there are
many reasons for a to be true while b would be false. In this case,
it's because OpenBSD -requires- completely open hardware, whereas the
documentation for PegII's gigEthernet is only available under NDA.
This violates one of OBSD's main tenets, therefore. no OBSD on PegII.

Someone should be talking to the Net-, Free-, or DragonFlyBSD people,
instead.

"I am an A1 owner, and I approve this message" [/cliche]
Notes about my presentation at the Alchimie 4 show : Comment 131 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 14:41 GMT
In reply to Comment 120 (Stéphane Guillard):
Of course I noticed that you're running a special version. That's why
I asked you to run the normal version, duh!

Unless OS4 does some really weird stuff, the emulated CPU meter should
be fine. I find it difficult to imagine an emulator that wouldn't
yield linear results for a program of this type (short loop with a
very large number of iterations). Your talk about "68k CPU%" is
nonsense though - like there would be one rating for "free 68k cycles"
and another one for "free PPC cycles".
Notes about my presentation at the Alchimie 4 show : Comment 132 of 427ANN.lu
Posted by Christophe Decanini on 29-Sep-2004 14:42 GMT
In reply to Comment 125 (itix):
No, Mike Bouma post as anonymous. Look at this thread you may find him.
Notes about my presentation at the Alchimie 4 show : Comment 133 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 14:46 GMT
In reply to Comment 130 (syrtran):
I believe the "completely open hardware" talk is nonsense, though. It
seems to me that OpenBSD runs on many platforms featuring less than
completely open hardware. It's another matter that OpenBSD may not
*support* that particular hardware.
Notes about my presentation at the Alchimie 4 show : Comment 134 of 427ANN.lu
Posted by Christophe Decanini on 29-Sep-2004 14:51 GMT
In reply to Comment 127 (Johan Rönnblom):
You are right but I didn't pay attention to the missing "open".
It is very misleading in the same sentence as openbsd is a bsd.
Notes about my presentation at the Alchimie 4 show : Comment 135 of 427ANN.lu
Posted by priest on 29-Sep-2004 14:53 GMT
In reply to Comment 124 (itix):
Stop stirring shit, it helps.

"Yes, as I see this VIA IDE vs eth problem is not Articia S problem but just a design flaw no-one noticed earlier."

Right because there were no UDMA drivers for AOS4 and because Linux drivers cause data corruption.

Now AOS4 UDMA drivers work flawlesly (knocking on the wood) but motherboard IDE DMA collides with the motherboard ethernet causing *????

"But nevertheless both Peg1 and A1 got broken on-board IDE and need HW fix to run flawlessly."

Not exactly.
Peg1 got "broken" northbridge (at least 2 bugs).
A1 got "broken" northbridge (1-2 bugs), "broken" southbridge (2bugs) and "broken" motherboard (this latest one).

The problem: GARDA gave you guys too few bugs ! ;-)

"Genesi got shit because of ArticiaS bugs, Eyetech is getting shit because of mobo bugs"

Let's not stirr it.

ArtisiaS has serious "issues". Or serious sh*t shall we say.
Newer revision improved memory compatibility and fixed some of the issues.
Even the latest ArtisiaS needs SPECIAL SW support to work without data corruption.
VIA IDE needed a separate SW fix.
Motherboards needs ???? replacement ????


People have said that they have fixed this UDMA-ethernet collision on some boards somehow (jump wires?).

Let's just wait for more news.
Notes about my presentation at the Alchimie 4 show : Comment 136 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 14:54 GMT
In reply to Comment 134 (Christophe Decanini):
Yep, it's very similar to the way Stéphane mentioned "VIA" in between
the other words. Very easy to miss, *even* if you're not a blue troll.
Notes about my presentation at the Alchimie 4 show : Comment 137 of 427ANN.lu
Posted by hooligan/dcs on 29-Sep-2004 14:55 GMT
In reply to Comment 107 (priest):
>While waiting for Eyetech response on how to fix this VIA UDMA - Ethernet collision ...

Alan is packing tin and jumpwires in envelopes, licking stamps and writing addresses. They should reach customers any week now ;)
Notes about my presentation at the Alchimie 4 show : Comment 138 of 427ANN.lu
Posted by AmiTroll on 29-Sep-2004 14:57 GMT
In reply to Comment 130 (syrtran):
>"I am an A1 owner, and i approve this message."[/cliche]

What is your stance on international trollism?
Notes about my presentation at the Alchimie 4 show : Comment 139 of 427ANN.lu
Posted by scheduler on 29-Sep-2004 14:57 GMT
In reply to Comment 73 (Thomas Frieden):
@Thomas Frieden

ScsiSpeed / DiskSpeed doesn't patch exec. See the source code if you don't believe it, it is available at: http://www.aminet.net/pub/aminet/disk/moni/DiskSpeed42.lha

Next excuse, thank you.
Notes about my presentation at the Alchimie 4 show : Comment 140 of 427ANN.lu
Posted by priest on 29-Sep-2004 14:57 GMT
In reply to Comment 129 (Johan Rönnblom):
YAWN!

Are you implying that Peg1 with April2 still corrupts data ?
Notes about my presentation at the Alchimie 4 show : Comment 141 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 14:58 GMT
In reply to Comment 135 (priest):
Sounds reasonable. Btw, I really don't mind if people want to claim
that in their opinion, the A1 has working UDMA now as they have tested
a driver and it seems to work.

However, I think otherwise, and until someone manages to release a
working driver for *any* platform, I think I have a very good case. So
if people try to claim that the UDMA works "for a fact", I will
disagree with them. For now, this is certainly very much in question.
Notes about my presentation at the Alchimie 4 show : Comment 142 of 427ANN.lu
Posted by itix on 29-Sep-2004 14:59 GMT
In reply to Comment 135 (priest):
At least problems are now identified and people know how to deal with them. It is not my problem so better stop it from my side here.
Notes about my presentation at the Alchimie 4 show : Comment 143 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 15:00 GMT
In reply to Comment 140 (priest):
No, I'm arguing that the supposed "proof" (SGs presentation) is not a
proof at all, as it would then also mean that the Peg 1 was bug-free
even without the April patch. And since everyone agree that the Peg 1
had a problem before the April patch, that must be wrong. Ergo: This
is not a valid proof, at all.
Notes about my presentation at the Alchimie 4 show : Comment 144 of 427ANN.lu
Posted by Fabio Alemagna on 29-Sep-2004 15:00 GMT
In reply to Comment 126 (Thomas Frieden):
Gee, for someone who doesn't overreact that's quite a rant. :-D

> > You may have noticed I "popped up" in the other thread too, in your defense.
>
> Didn't notice... which thread do you mean ?

Come on, do your homework. Ask your brother. Either you or him replied to one of my posts in that thread.

> > Perhaps those SW actually used undocumented features, or simply relied on
> > side effects of bugs which couldn't really be kept.
>
> Perhaps ?

You ask me? Do you want me to say "I'm certain that..." even if I am not? Funny, I was giving your the benefit of the doubt by saying "perhaps", but if you want I can say "I'm certain that...", which however would completely rule out your reasons for breaking compatibility.


> > Care to mention which SW you're talking about, and which are the
> > incompatibilities you're thinking of?
>
> Are you kidding ?

No I'm not. You are so sure that the previous releases have broken compatibility in places where it was documented that certain things would work in a certain way, that you surely must know what are those places, and what SW is that got broken. You made the claim, be ready to prove it.

> It's well known that a lot of software broke between 1.3 and 2.0. I don't need
> to go into details.

"It' well known"? :-) So, if it' well known, you know too. Hence, please, give some details.

> > You know, it helps talking about real stuff.
>
> LOL. Yes, that's true.

Glad you agree, too bad you stop there (at the agreement, that is).

> > Really? Is that a law or something?
>
> Nonsense.

You really can't help writing at least once "nonsense" or "troll" in your posts, right? You do realize you come off as pretty arrogant, don't you?

> > Let's say things differently - and
> > properly - instead: "When moving from one revision to the next, some people
> > think it's ok to lose some compatibility".

> Ok, if you see it that way.

No, it is that way. That, or you admit that breaking compatibility is something that is enforced by some superior authorities. It's either an opinion - as in my sentence above - or a fact. So, which is which?

> The fact remains that for some things, it's worth dropping compatibility, and
> that is the case here.

Whether that is the case here is debatable, but I won't debate it.

> > You may want to notice that I've refrained from commenting on whether the
> > fact AOS4 breaks compatibility is a good or a bad thing, nonetheless you've
> > jumped off your chair
>
> I didn't jump off my chair. You overrate yourself.

Dude, you do a bad job at looking calm :-) It's not a matter of "overrating" myself, I would overrate myself if I thought you stayed calm just because you were talking to me. But you got... err... blue in the face :-D

> > as if I had said anything negative about it, up to the
> > point of indirectly (as cowardy is too strong, it seems) calling me a troll.
>
> I also didn't call you a troll.

Yes yes, sure :-)

> But if you want to, OK: You are a troll. Every time OS4 is mentioned, you jump
> off your chair.

No, I didn't want to. But thanks for confirming that you, indeed, think I'm troll, just for stating what is a fact: AmigaOS4 breaks compatibility with the previous releases in many places.

> See ? Better now ?

I dunno, you tell me. Do you feel better now that you took of your mask? :-)


> > I see. So, you're saying that, say, MOS has not and is not "moving forward"?

> Here we go. That's your typical way of "arguing". Where did I say MOS ? Where
> did I say AROS ? Right. I didn't.

Right, you didn't. In fact, that's what I said too, didn't I? Oh, look, you cut exactly the part where I say that you didn't talk about MOS? How's that, pal? ;-)

> > I'm kind of tired of your habit of not talking about anything concrete, so
> > please, answer - politely, if possible - to my question.
>
> Why should I be polite to you when you insult me, calling me a coward ?

In other words, you don't know what to answer.

> And it's really great to see you accusing _ME_ of not talking about anything
> concrete. You NEVER talk about anything concrete. You always gloss or skip
> over parts where you would have to be concrete. You don't answer questions
> asked to _YOU_ when you don't want to asnwer them, but keep on bringing in
> stuff that's completely irrelevant to the point, like you just did by bringing
> MOS into the picture. That's why I think you are a troll (See, I call you a
> troll again, I must be getting bold).

Given that you completely misunderstood the MOS point, care to mention whicha re those questions I'm asked and which I don't answer to, which you are referring to? Failure to do so, will confirm that you, indeed, only talk about hot air :-)

> But to answer your question: When I said "moving forward", I wasn't even
> thinking about anything else but OS3->OS4.

I am not in your brain (thankfully), how am I supposed to know that a general statement like "this is the price of moving forward" was only meant to be applied to AOS4?

> It's my opinion that moving forward involves dropping compatibility.

Ok, now we know what's the level of clueness of the people working on AOS4.

> For example, we're currently working on a nearly constant time memory
> allocation system. It will definitely be incompatible to some programs, but
> heck, reducing memory fragmentation and brining down allocation time is worth > the trouble... Oh, of course in my opinion... No law involved...

As said, it may be true or may be not true that in that specific case breaking compatibility is worth it, however your general statement that breaking compatibility is a conditio sine qua non "moving forward" is not possible is completely bogus, regardless of whether that's your opinion: at worst, you've got a bogus opinion. Of course, you can regard that as my (possibily bogus, if you like) opinion :-)

> > Of course, we all know that the only useful comments are the ones that
> > support you, your product and your company ;-)

> Useful ? I said pointless... meaning it was not to the point.

Sorry, I couldn't find any other term, off the top of my head, to mean the oppisite of "pointless". I'll let you reword the above sentence to fit this requirement :-)

> Regarding the pointlessness of anti-OS4 postings: It should go without saying
> that I'm bias ;-) However, I think I was never really unreasonable.

Oh, but I also think I have never been unreasonable!

Bah, too much to reply too still, and I've got more important stuff to do. Cya.
Notes about my presentation at the Alchimie 4 show : Comment 145 of 427ANN.lu
Posted by priest on 29-Sep-2004 15:11 GMT
In reply to Comment 141 (Johan Rönnblom):
Reasonable comment.

After all ... I see this UDMA-Ethernet collision as a FATAL bug.
It makes one (almost) angry that they missed it.

No, forget the "almost" part.
Notes about my presentation at the Alchimie 4 show : Comment 146 of 427ANN.lu
Posted by Christophe Decanini on 29-Sep-2004 15:17 GMT
In reply to Comment 130 (syrtran):
Yes I didn't see that "open" was missing in the second sentence.
Yes I didn't specify onboard UDMA but I thought that it was obvious that I was speaking about the onboard one.

The point is that all these discusions are worthless.
Calling me a troll or saying "you re wrong" without any explanation shows how some people will catch every single detail to prove someone else wrong when they would/could have answered along the lines "Right, speaking of the onboard UDMA driver forever does not make sense when it can not be used with ethernet".

Were all people who didn't specify "on board" when speaking about IDE, sound, USB ... called
trolls ?

Now we can stop the rhetoric talks and speak about the real problems:

How does it affects A1 users ?
How will Eyetech/ vendors assist the users ?
When will working drivers be available ?

It looks to me that every time Hyperion Eyetech stated "it is working" "there is no bug" it was to find out later there was some that were not yet discovered.
Software bugs are common and painless compared to these hardware bug that should have really have been addressed before selling 1000+ motherboards.
Notes about my presentation at the Alchimie 4 show : Comment 147 of 427ANN.lu
Posted by priest on 29-Sep-2004 15:18 GMT
In reply to Comment 137 (hooligan/dcs):
and the A1 repair shops rice like mushrooms at rainy autumn. ;)
Notes about my presentation at the Alchimie 4 show : Comment 148 of 427ANN.lu
Posted by Johan Rönnblom on 29-Sep-2004 15:19 GMT
In reply to Comment 145 (priest):
I'm sure they missed it because those drivers didn't exist, so it was
untested. I think the major cause for the problem here is the
readiness of some to take a lot of things for granted, even though it
has not been tested. Under such circumstances, problems should be
expected. Hence the long delays, problems which weren't supposed to be
there, etc.
Notes about my presentation at the Alchimie 4 show : Comment 149 of 427ANN.lu
Posted by Bernie Meyer on 29-Sep-2004 15:23 GMT
In reply to Comment 145 (priest):
One does have to wonder how such a detail was left out from the earlier "Yes, UDMA works just fine, we just don't include it in the Prerelease because we want to be 100% sure" statements. After all, surely *someone* was using a hard disk on that VIA UDMA driver to feed data to a network...BTW, well done on the Nalle Puh port. Any chance you might release the sources to someone who isn't, strictly speaking, entitled to them? I'd love to see how you hook into the 68k emulation for that one :)
Notes about my presentation at the Alchimie 4 show : Comment 150 of 427ANN.lu
Posted by Christophe Decanini on 29-Sep-2004 15:23 GMT
In reply to Comment 145 (priest):
"After all ... I see this UDMA-Ethernet collision as a FATAL bug. "

You're wrong ;)
Anonymous, there are 427 items in your selection (but only 327 shown due to limitation) [1 - 50] [51 - 100] [101 - 150] [151 - 200] [201 - 250] [251 - 300] [301 - 350] [351 - 400] [401 - 427]
Back to Top