27-Apr-2024 16:03 GMT.		 UNDER CONSTRUCTION
Anonymous, there are 115 items in your selection (but only 65 shown due to limitation) [1 - 50] [51 - 100] [101 - 115]
[Events] Nice Essen ReportANN.lu
Posted on 04-Jun-2004 13:09 GMT by Alkis Tsapanidis (Edited on 2004-06-04 19:56:11 GMT by Christophe Decanini)115 comments
View flat
View list
Here you can find a very nice and informative report of the Essen AmigaOS4 Event.
Nice Essen Report : Comment 51 of 115ANN.lu
Posted by Anonymous on 04-Jun-2004 16:28 GMT
In reply to Comment 45 (Sammy Nordström):
PCB: yes, I think the Pegasos one cleaner and more professional, just compare it.
Especially, there is no connector proctection on current AOne...

the Aone is a devellopement board for the ArticiaS named TeronCX ...

About the IDE driver, well I'm not especially speaking about Hyperion, but people involved in the AOne/Teron: Hyperion/May/Eyetech ...

Bye
Nice Essen Report : Comment 52 of 115ANN.lu
Posted by Sammy Nordström on 04-Jun-2004 16:32 GMT
In reply to Comment 40 (Anonymous):
To counter-argue your comment about the IDE driver for AmigaOS4:

If it takes the MorphOS Team more than 7 years to make a PPC native TCP/IP stack, it's going to take *centuries* before your Pegasos will be fully usable.

Now, the above statement is of course so full of erranous logic that it's not even funny, and I hope that this might make you realize a thing or two about the flaws in your own statements about the IDE drivers for AmigaOS4.
Nice Essen Report : Comment 53 of 115ANN.lu
Posted by Anonymous on 04-Jun-2004 16:36 GMT
In reply to Comment 52 (Sammy Nordström):
I was not espeically about the OS4 crap. I guess the "workround" in the IDE driver, is of the same kind that the one is the Linux one.
Nice Essen Report : Comment 54 of 115ANN.lu
Posted by Alkis Tsapanidis on 04-Jun-2004 16:40 GMT
In reply to Comment 52 (Sammy Nordström):
I do not support this guy's statement about the IDE driver but I can tell you that
1) MorphOS has been in development for something more than 5 years.
2) A TCP/IP stack wasn't even in development. The core (AmiTCP5PPC) has been
working for ages, there were no highlevel parts.
3) It would use Roadshow but then Olaf signed the exclusive PPC agreement with
Hyperion.

...

4) You still haven't replied to my counterarguements in the other thread...
You were mistaked and did not admit it.
Nice Essen Report : Comment 55 of 115ANN.lu
Posted by Anonymous by Design on 04-Jun-2004 16:42 GMT
I get far more HATE mail from Ben Hermans.
Nice Essen Report : Comment 56 of 115ANN.lu
Posted by Sammy Nordström on 04-Jun-2004 16:43 GMT
In reply to Comment 51 (Anonymous):
>PCB: yes, I think the Pegasos one cleaner and more professional, just compare
>it. Especially, there is no connector proctection on current AOne...

...and this makes the Pegasos PCB more clean because...?

>the Aone is a devellopement board for the ArticiaS named TeronCX ...

Nope. The TeronCX is the development board for the ArticiaS, not the AmigaOne.

>About the IDE driver, well I'm not especially speaking about Hyperion, but
>people involved in the AOne/Teron: Hyperion/May/Eyetech ...

Yes, it's very common that OS developers needs to be in contact with the hardware developers when writing hardware drivers...

Again, what makes you think they've spent 2 years on nothing else but writing IDE drivers for the AmigaOne?
Nice Essen Report : Comment 57 of 115ANN.lu
Posted by MIKE on 04-Jun-2004 16:48 GMT
In reply to Comment 56 (Sammy Nordström):
No one thinks you've written any IDE drivers for the AmigaONE samface. If you could manage to write a working arexx script, I think most folks would be shocked.
Nice Essen Report : Comment 58 of 115ANN.lu
Posted by Sammy Nordström on 04-Jun-2004 16:49 GMT
In reply to Comment 54 (Alkis Tsapanidis):
MorphOS has been in development since 1997.

As for the rest, I was perfectly aware of the erranous logic in the statement I made about the MorphOS TCP/IP stack. The point was to show the erranous logic behind the claim that if it takes two years to write an IDE driver, it will take 20 years before the system becomes usable. It is the same erranous logic based on the same kind of misconceptions. See?
Nice Essen Report : Comment 59 of 115ANN.lu
Posted by Sammy Nordström on 04-Jun-2004 16:52 GMT
In reply to Comment 57 (MIKE):
Would it surprise you if I said that I know basic, C/C++, PHP, Perl and MySQL programming? Well, I do mostly PHP and MySQL these days, www.mindrelease.net beeing my current project.

But you're correct, I have no experience from writing hardware drivers and neither did I claim that I would. Besides, what's that go to do with anything? We're not talking about anything technical here, now are we?
Nice Essen Report : Comment 60 of 115ANN.lu
Posted by Anonymous on 04-Jun-2004 16:53 GMT
In reply to Comment 56 (Sammy Nordström):
Well, the general PCB design/quality. Of course, it's a bit subjective ...

-
Well, usually when you sell a motherboard to the end-user, you provide _at least_ the minimal software needed. Here, I consider the _minimal_ software has:

Working Linux (IDE/Network/GFX). Obviously, the IDE was not working (has everytinhg playing with the DMA)... So I hope MAY/Eyetech/Hyperion worked night and day to minize the delay to get usable drivers to users... But maybe they noticed it's impossible so they gave up (Remember MAI remove the DMA feature of the ArticiaS on their web site)

About what about poor SE users with the AriticaS memory trash ... They will never get a usable computer...
-
And yes, I'm sorry the AOne is a TeronCX with and updated BIOS.
Nice Essen Report : Comment 61 of 115ANN.lu
Posted by Anonymous by Design on 04-Jun-2004 16:57 GMT
Now why would Petro not be happy will Bill Buck, after all Bill has done for him? Maybe Petro is unhappy with his 'present'. Was it not Petro who suggested Bill go ahead with the lawsuit, maybe he changed his mind later?

Either way, its all Ben's fault the current mess we are in.

http://www.ping.de/sites/tarbos/petrot_alanr.jpg
Not seen in that picture is Alan sliding Petro some 'Red' money.

Alan "Ehy Petro this is your *insert your favourite british curse word* money. Do as you like wit it!"

Petro " *sigh* Where's the beer? Alan you promised me there would be some hottie chickie-pies, where are they? This party sucks! *sigh*"
Nice Essen Report : Comment 62 of 115ANN.lu
Posted by Alkis Tsapanidis on 04-Jun-2004 17:02 GMT
In reply to Comment 58 (Sammy Nordström):
In 1997 PowerUP was still in development... MorphOS' development started in '99.

I am fully aware of what you meant, I just replied to an arguement used ALL the
time about MorphOS' usuability.
Nice Essen Report : Comment 63 of 115ANN.lu
Posted by Sammy Nordström on 04-Jun-2004 17:22 GMT
In reply to Comment 60 (Anonymous):
IDE always was and still is working. You're talking about DMA, a problem they have fixed by writing proper drivers. The standard linux drivers didn't work because the chipset does not comply with certain PC standards. Some people prefer to call it a bug, some people prefer to call it a feature. Who prefers what is pretty easy to guess. As for me, I think innovative hardware designs has always had its drawbacks, not beeing "standard" is one of them.

I'd say it's about time you MorphOS/Pegasos supporters drop the Articia bug argument now. We've been over this so many times before and you need to realize that this coin has two sides and that you cannot force anyone see things your way. If you don't own an AmigaOne, it's not even your problem to begin with, now is it?
Nice Essen Report : Comment 64 of 115ANN.lu
Posted by Anonymous on 04-Jun-2004 17:39 GMT
In reply to Comment 63 (Sammy Nordström):
"IDE always was and still is working. You're talking about DMA, a problem they have fixed by writing proper drivers. The standard linux drivers didn't work because the chipset does not comply with certain PC standards. Some people prefer to call it a bug, some people prefer to call it a feature. Who prefers what is pretty easy to guess. As for me, I think innovative hardware designs has always had its drawbacks, not beeing "standard" is one of them. "

You are kidding ?
If you really think that you are writting, then I'm really sorry for you.

Ok, let's say it's a feature, and etc... Please, submit the VIA IDE driver patch to the Linux guys. And see what they think about it !

Bye
Nice Essen Report : Comment 65 of 115ANN.lu
Posted by Klaus on 04-Jun-2004 18:11 GMT
"Yeah, I remember some slight a chart with slight differences as well. Other reasons for differences could be that the new DMA drivers allow "simultaneous memory access by the CPU and a PCI bus master" on the AmigaOne and the Pegasos1 disabled this feature and that the Pegasos2 does not have 1X AGP bandwith "Usually where an AGP slot is electrically implemented on a PCI-X bus this only runs with the 66 MHz PCI (not PCI-X) protocol and is therefore significantly slower than even single speed AGP."


But even this "slower" AGP Port on Peg2 is at least as fast as the AGP x2 on Peg1. So what?
Nice Essen Report : Comment 66 of 115ANN.lu
Posted by ehaines on 04-Jun-2004 18:12 GMT
In reply to Comment 60 (Anonymous):
> About what about poor SE users with the AriticaS memory trash ... They will never get a usable computer...

A lie, of course, seeing as how I've had a plenty usable SE for over a year now and somehow none of my files seem to be corrupted. Nor does it ever randomly lock/crash/etc.

Strange how those most critical of this haven't actually used it....
Nice Essen Report : Comment 67 of 115ANN.lu
Posted by Anonymous on 04-Jun-2004 18:40 GMT
In reply to Comment 66 (ehaines):
Then stress it, enable DMA, make endless md5sum loop ...
Nice Essen Report : Comment 68 of 115ANN.lu
Posted by Amon_Re on 04-Jun-2004 19:47 GMT
In reply to Comment 50 (Rich Woods):
Well, Petro isn't happy with Ainc neighter, he feels the community has been led down by people like Bill Mcewen & Bill Buck, he said something in the line of "everyone is doing what they want, not what is good for the community".

He's disapointed about the whole mess, Petro always cared about the community, and he feels that this is missing with some of the big players, not just from BBRV, but he's not fond of BBRV at all.

Cheers
Nice Essen Report : Comment 69 of 115ANN.lu
Posted by Some Farker on 04-Jun-2004 20:47 GMT
In reply to Comment 67 (Anonymous):
There are lies, damn lies, and anonymous pegasus using trolls. How many promisary notes has Bill Buck sent you to spread FUD? (He's broke, y'know - you'll never get paid.)
Nice Essen Report : Comment 70 of 115ANN.lu
Posted by jkdsteve2@hotmail.com on 04-Jun-2004 23:59 GMT
In reply to Comment 63 (Sammy Nordström):
I'd say it hasn't been satisfactorily answered. If it was 'feature'
why the hell did it take MAI a year or more to 'admit' it? Why stand
there dumbstruck in front of a bunch of engineers wanting to buy your
product in the 1,000s but they can't because of problems with DMA data
transfers???

Just doesn't add up...and this is well before the April crap
rumourmongering and even Genesi etc.

The rumours will stop when the explanation matches the bloody
evidence.
Nice Essen Report : Comment 71 of 115ANN.lu
Posted by itix on 05-Jun-2004 00:22 GMT
In reply to Comment 52 (Sammy Nordström):
Actually they dont make TCP/IP stack themselves, they port AmiTCP. Nevertheless Roadshow was going to be ported for MorphOS but Hyperion blocked this with their exclusive contract.
Nice Essen Report : Comment 72 of 115ANN.lu
Posted by itix on 05-Jun-2004 00:26 GMT
In reply to Comment 66 (ehaines):
Start using Linux or Mac and you see what is the prob. Funny that Linux works correctly on every other PPC mobo.
Nice Essen Report : Comment 73 of 115ANN.lu
Posted by itix on 05-Jun-2004 00:29 GMT
In reply to Comment 68 (Amon_Re):
But since when Petro did anything useful? Is he different from Bill Buck or Ben Hermans other than that he says nice words?
Nice Essen Report : Comment 74 of 115ANN.lu
Posted by greenboy on 05-Jun-2004 00:56 GMT
In reply to Comment 68 (Amon_Re):
> Amon_Re :
>Petro always cared about the community, and he feels that this is missing with some of the big players, not just from BBRV, but he's not fond of BBRV at all.

I have to wonder if this isn't more spun-up anti-Buck rhetoric coming from other mouths entirely, and its just convenient for your I Hate Buck crusade that you've been doing to death forever and a day. Or if the context is being stretched to provide the Reds with their fave hate doll. Sure wouldn't be the first time.

As far as I've known Petro and Bill have been friends for a long time. They might be frank types, but they aren't likely to break their ties over the dumbasses in the community.

Amon Re, I have to say in many ways I think you're a pretty decent fellow. But that spew you always do about Buck does you no credit after so long.
Nice Essen Report : Comment 75 of 115ANN.lu
Posted by ehaines on 05-Jun-2004 01:03 GMT
In reply to Comment 67 (Anonymous):
> Then stress it, enable DMA, make endless md5sum loop ...

I do have DMA enabled...on my machine, for whatever reason, MOL doesn't really work if it's disabled. As for md5sums, I can't be bothered frankly...if it runs in normal use for over a year with no bizarre behavior or lost data, that's all I care about.
Nice Essen Report : Comment 76 of 115ANN.lu
Posted by itix on 05-Jun-2004 01:07 GMT
In reply to Comment 75 (ehaines):
FYI you can run pre-April Peg1 board for years without problems, but in one day everything blows up.
Nice Essen Report : Comment 77 of 115ANN.lu
Posted by ehaines on 05-Jun-2004 01:09 GMT
In reply to Comment 72 (itix):
> Start using Linux or Mac and you see what is the prob. Funny that Linux works correctly on every other PPC mobo.

What are you talking about? As some are fond of pointing out, I can only run Linux on it so far, and probably for a few weeks yet. Plus I run MOL. It's been my primary machine for months now (since my old Amiga stopped working right). Sorry to disappoint you, but it doesn't continuously bomb or freeze or lose data or other such made-up stuff. It's not the best computer ever made, but it's plenty usable.

Again, those most critical don't actually have one or have never seen or used one. All they have are agendas and axes which seem to need grinding. Pathetic really.
Nice Essen Report : Comment 78 of 115ANN.lu
Posted by gary_c on 05-Jun-2004 03:13 GMT
In reply to Comment 17 (DaveP):
Yes, I read it rather as the reviewer reporting what was said, rather than editorialising.

This isn't how reporting works. The reporter's first choice is to get a direct quote ("Alan said, 'Bill Buck is evil.'"). Next best is to paraphrase, still attributing ("Petro agreed that Bill Buck is evil.") Whenever we see unattributed paraphrases, then the reporter is giving us his/her interpretation of statements and events. This is still valid journalism, but is subject to greater scrutiny because of the lack of clear sources. When you cross the line from reporting, strictly speaking, to interpretation and summation, that's getting into editorial content. Given the article's overall tone, which is clearly the intention to show some parties in a positive light and others negatively, I have to question how valid any particular unattributed statement could be.

But I understand and empathise with the sensitivities expressed in this thread.

Well, apparently the writer wasn't too concerned with accuracy. Must have been fun to write, but the more the writer indulges his biases, the more worthless the article is.

-- gary_c
Nice Essen Report : Comment 79 of 115ANN.lu
Posted by Sammy Nordström on 05-Jun-2004 04:20 GMT
In reply to Comment 70 (jkdsteve2@hotmail.com):
>The rumours will stop when the explanation matches the bloody
>evidence.

Alright. Let's try. Give me the evidence (no, not claims, statements, etc. give me cold hard factual proof) and let's see what we can explain and what we can't.
Nice Essen Report : Comment 80 of 115ANN.lu
Posted by DaveP on 05-Jun-2004 04:32 GMT
In reply to Comment 78 (gary_c):
Gary

You know thats the best written comment you have ever made.

Dave.
Nice Essen Report : Comment 81 of 115ANN.lu
Posted by Don Cox on 05-Jun-2004 04:59 GMT
In reply to Comment 67 (Anonymous):
"Then stress it, enable DMA, make endless md5sum loop ..."

What would be the point of him enabling DMA in Linux? Everybody knows that gives trouble on the AmigaOnes. The argument is about why, and about whether AOS4 has the same problem(s).

The argument will be settled only when somebody finds and demonstrates a solution, and everyone can see how that solution works.

No point in discussing it now. Wait at least until all the A1 users have an IDE driver that uses DMA.
Nice Essen Report : Comment 82 of 115ANN.lu
Posted by Don Cox on 05-Jun-2004 05:04 GMT
In reply to Comment 70 (jkdsteve2@hotmail.com):
"Why stand
there dumbstruck in front of a bunch of engineers wanting to buy your
product in the 1,000s but they can't because of problems with DMA data
transfers???

Just doesn't add up...and this is well before the April crap
rumourmongering and even Genesi etc."

Could be that MAI have nobody who is competent in Linux internals. After all, they are a chip design company. They probably didn't expect it to be a problem.


"The rumours will stop when the explanation matches the bloody
evidence."

Exactly.
Nice Essen Report : Comment 83 of 115ANN.lu
Posted by Sammy Nordström on 05-Jun-2004 05:53 GMT
In reply to Comment 81 (Don Cox):
>The argument will be settled only when somebody finds and demonstrates a
>solution, and everyone can see how that solution works.

http://amigaworld.net/modules/news/article.php?storyid=1362
Nice Essen Report : Comment 84 of 115ANN.lu
Posted by jkdsteve2@hotmail.com on 05-Jun-2004 06:51 GMT
In reply to Comment 79 (Sammy Nordström):
Can't be done..and won't be done. Even if I had the hardware,
bustraces etc. What would be the point? The explanation still doesn't
make sense...or at least if it does, there's something very wrong at
MA:

I repeat again, since you missed the point, rephrasing and expanding:

What did the bplan engineers have to gain by making false claims
about an important component of a product they already spent many
months on and intended to buy in the 1,000s? There's simply no motive
here.

and again...

Why did MAI not understand what was been shown to them?

We are talking about engineers here, with logical behaviour,
analytical mindsets. "Here are the symptoms. Here's how to trigger
them. There seems to be a problem or bug."

The answer from MAI should not be ignorance or denial (or no answer
at all...I don't know which it was..I wasn't there.) - it's research
and, if they know their own chip inside out someone *MUST SURELY HAVE
SPOTTED* the relation between DMA I/O and this 'PCI coherence
feature'

It might seem like I waste my time raking over the past but I really
want to know because 2+2 are making 5 here!
Nice Essen Report : Comment 85 of 115ANN.lu
Posted by Sammy Nordström on 05-Jun-2004 07:13 GMT
In reply to Comment 84 (jkdsteve2@hotmail.com):
>Can't be done..and won't be done.

With just claims and accusations without something factual to back it up, there is nothing to explain.

> Even if I had the hardware,
>bustraces etc. What would be the point?

It would give the accused party the ability to explain. For as long as they don't know what your accusations are based on, it's impossible for them to explain anything.

> The explanation still doesn't
>make sense...

Which explanation to which accusation?

>What did the bplan engineers have to gain by making false claims
>about an important component of a product they already spent many
>months on and intended to buy in the 1,000s? There's simply no motive
>here.

I'm sure it was not planned to turn out this way from the beginning, but I'm also quite sure that certain Genesi individuals would not hesitate to make use of such opportunity to throw mud at the competition...

>Why did MAI not understand what was been shown to them?

What is it that they didn't understand and how do you know that they didn't understand it?

>We are talking about engineers here, with logical behaviour,
>analytical mindsets. "Here are the symptoms. Here's how to trigger
>them. There seems to be a problem or bug."

Well, it's not like there is no politics involved...

>The answer from MAI should not be ignorance or denial (or no answer
>at all...I don't know which it was..I wasn't there.) - it's research
>and, if they know their own chip inside out someone *MUST SURELY HAVE
>SPOTTED* the relation between DMA I/O and this 'PCI coherence
>feature'

How do you know the answer from MAI?

>It might seem like I waste my time raking over the past but I really
>want to know because 2+2 are making 5 here!

2 baseless accusations doesn't make one fact either, you know...
Nice Essen Report : Comment 86 of 115ANN.lu
Posted by JKD on 05-Jun-2004 07:37 GMT
In reply to Comment 85 (Sammy Nordström):
Sammy sammy sammy....you really are a piece of work.

You realized that I was speaking historically right? I have to be sure because English may not be your native language.

I'm stating an opinion, all the accusations referenced in my text are paraphrasing historical statements by 1st, 2nd and 3rd parties to this issue with Articia and MAI. I, personally am accusing no one with my line of reasoning..by inference I am repeating someone elses allegations, that is all.

Now get off the fence and start debating instead of talking semantics. obsfucating and skirting the bloody issue at hand...even better tell us what you believe about my line of reasoning.

I'll give you a proper answer to your semantics and blustering when I have more time.

Steve
Nice Essen Report : Comment 87 of 115ANN.lu
Posted by Anonymous on 05-Jun-2004 07:49 GMT
In reply to Comment 83 (Sammy Nordström):
Bring, the IDE device guy here, so we can discuss about the workaround and PROVE that the ArticiaS is pure crap, that it takes ages to make a proper driver (which by the way is SLOWER than on Pegasos...), and that it can't be done for others devices...
Nice Essen Report : Comment 88 of 115ANN.lu
Posted by JKD on 05-Jun-2004 07:50 GMT
In reply to Comment 85 (Sammy Nordström):
>>Can't be done..and won't be done.
>
>With just claims and accusations without something factual to back it up, there is nothing to explain.

Do you believe that bPlan presented no evidence to MAI?

>> Even if I had the hardware,
>>bustraces etc. What would be the point?
>
>It would give the accused party the ability to explain. For as long as they don't know what your
>accusations are based on, it's impossible for them to explain anything.

I am not making accusations, merely trying to arrive at a logical conclusion based on a series of past events.

>> The explanation still doesn't
>>make sense...
>
>Which explanation to which accusation?

You know, I hope you really are smarter than that - or is it because you lose something in translation?

>>What did the bplan engineers have to gain by making false claims
>>about an important component of a product they already spent many
>>months on and intended to buy in the 1,000s? There's simply no motive
>>here.
>
>I'm sure it was not planned to turn out this way from the beginning, but I'm also quite sure that
>certain Genesi individuals would not hesitate to make use of such opportunity to throw mud at the >competition...

later spin on events is irrelevant. I am attempting to stick to events at a given point intime..in fact some of this occurred even pre-Genesi

>>Why did MAI not understand what was been shown to them?
>
>What is it that they didn't understand and how do you know that they didn't understand it?

C'mon, now you're pulling my leg (means joking) right? Why did they not simply say to bPlan whether immediately or after some investigation that the results they are seeing are correct because they have accounted for feature XXX in the chipset.

>>We are talking about engineers here, with logical behaviour,
>>analytical mindsets. "Here are the symptoms. Here's how to trigger
>>them. There seems to be a problem or bug."
>
>Well, it's not like there is no politics involved...

In the timeframe this ocurred, the politics were irrelevant and agian, it's not cntral to my arguement and you did not answer the question I posed (but I suppose I should be used to that by now!)

>>The answer from MAI should not be ignorance or denial (or no answer
>>at all...I don't know which it was..I wasn't there.) - it's research
>>and, if they know their own chip inside out someone *MUST SURELY HAVE
>>SPOTTED* the relation between DMA I/O and this 'PCI coherence
>>feature'
>
>How do you know the answer from MAI?

I don't (and neither do you) - care to elaborate on your line of reasoning? Given that bPlan apparently failed to receive any input on the problem that would help them fix it, I am inferring that MAI were negligent and acted in a way that commercially made no sense.

>>It might seem like I waste my time raking over the past but I really
>>want to know because 2+2 are making 5 here!
>
>2 baseless accusations doesn't make one fact either, you know...

Pointless comment, what I said was merely a common expression to say that the sum of the parts does not equal the whole, quit boring everyone with your 'wit', get off the darned fence and use less 'forked-tongue'

Steve
Nice Essen Report : Comment 89 of 115ANN.lu
Posted by Sammy Nordström on 05-Jun-2004 08:05 GMT
In reply to Comment 86 (JKD):
My point was exactly what you just said; you are saying things that you have only heard from a rather biased source and then use it as the basis for your conclusion that the accused party's explanation doesn't make sense. How could you possibly judge wether the explanation makes sense or not when you don't even know what they are trying to explain?

So, in order to have a constructive debate regarding wether the explanations makes sense or not, we have to sort out which claims are valid or not. Simply refering to the accusations made by a rather biased source of information will not do.
Nice Essen Report : Comment 90 of 115ANN.lu
Posted by Sammy Nordström on 05-Jun-2004 08:24 GMT
In reply to Comment 88 (JKD):
>>>Can't be done..and won't be done.
>>
>>With just claims and accusations without something factual to back it up,
>>there is nothing to explain.
>
>Do you believe that bPlan presented no evidence to MAI?

I merely trying to sort out what we know and what we don't in order to maintian an objective and constructive debate.

>>> Even if I had the hardware,
>>>bustraces etc. What would be the point?
>>
>>It would give the accused party the ability to explain. For as long as they
>>don't know what your accusations are based on, it's impossible for them to
>>explain anything.
>
>I am not making accusations, merely trying to arrive at a logical conclusion
>based on a series of past events.

The accusation is that they're explanations doesn't make sense, remember? My point is that we need to investigate those series of past events before we make any conclusions based on them.

>>>What did the bplan engineers have to gain by making false claims
>>>about an important component of a product they already spent many
>>>months on and intended to buy in the 1,000s? There's simply no motive
>>>here.
>>
>>I'm sure it was not planned to turn out this way from the beginning, but I'm
>>also quite sure that certain Genesi individuals would not hesitate to make
>>use of such opportunity to throw mud at the competition...
>
>later spin on events is irrelevant. I am attempting to stick to events at a >given point intime..in fact some of this occurred even pre-Genesi

We don't know what happened. All we know is the information that has been made public later on as a part of the spin on the events. See?

>>>Why did MAI not understand what was been shown to them?
>>
>>What is it that they didn't understand and how do you know that they didn't
>>understand it?
>
>C'mon, now you're pulling my leg (means joking) right? Why did they not simply
>say to bPlan whether immediately or after some investigation that the results
>they are seeing are correct because they have accounted for feature XXX in the
>chipset.

Maybe they did, maybe they didn't. Who knows? Do you? If so, please share something factual to back it up with.

>>>We are talking about engineers here, with logical behaviour,
>>>analytical mindsets. "Here are the symptoms. Here's how to trigger
>>>them. There seems to be a problem or bug."
>>
>>Well, it's not like there is no politics involved...
>
>In the timeframe this ocurred, the politics were irrelevant and agian, it's
>not cntral to my arguement and you did not answer the question I posed (but I
>suppose I should be used to that by now!)

The politics were *never* irrelevant. If they would, things would have turned out quite differently...

Furthermore, it's the time of when this information you base this on became available that is relevant, not the time it actually occured.

>>>The answer from MAI should not be ignorance or denial (or no answer
>>>at all...I don't know which it was..I wasn't there.) - it's research
>>>and, if they know their own chip inside out someone *MUST SURELY HAVE
>>>SPOTTED* the relation between DMA I/O and this 'PCI coherence
>>>feature'
>>
>>How do you know the answer from MAI?
>
>I don't (and neither do you) - care to elaborate on your line of reasoning?
>Given that bPlan apparently failed to receive any input on the problem that
>would help them fix it, I am inferring that MAI were negligent and acted in a
>way that commercially made no sense.

How can you possibly say that they were negligent when you at the same time admit that you don't know what their answer was? And what do you mean by "bPlan apparently failed to receive any input", care to share whatever it is that makes it apparent with us?

>>>It might seem like I waste my time raking over the past but I really
>>>want to know because 2+2 are making 5 here!
>>
>>2 baseless accusations doesn't make one fact either, you know...
>
>Pointless comment, what I said was merely a common expression to say that the
>sum of the parts does not equal the whole, quit boring everyone with
>your 'wit', get off the darned fence and use less 'forked-tongue'

No, it was far from "pointless". I think it sums up what I'm trying to tell you quite well.
Nice Essen Report : Comment 91 of 115ANN.lu
Posted by ehaines on 05-Jun-2004 09:22 GMT
In reply to Comment 81 (Don Cox):
> What would be the point of him enabling DMA in Linux?

Because, as I said, MOL does not bloody well work WITHOUT it enabled, that's why! (This is true for my AmigaOne; dunno about anyone else's). Plus, it's faster.

> Everybody knows that gives trouble on the AmigaOnes.

What everyone knows, actually, is that it seems to depend on the individual AmigaOne and probably other factors like memory, etc. There are others beside me who have it enabled and it doesn't cause failures.

But look, I back my stuff up. Which everyone should do anyway, AmigaOne or otherwise, because you never know when your harddrive will die or some massive power surge will strike or whatever. So if the ArticiaS really does make my computer blow up once every few years, I'll be out the 10 minutes it will take to restore from backup. Big freakin' deal. There are more important things to worry about, I'd say. Like paying developers for example. (Ha, sorry, couldn't resist....)
Nice Essen Report : Comment 92 of 115ANN.lu
Posted by Anonymous on 05-Jun-2004 09:32 GMT
In reply to Comment 63 (Sammy Nordström):
Spamface brainfarted:

> I'd say it's about time you MorphOS/Pegasos supporters ...

I think it's about time you fucking retarded 'red vs. blue' factionist babies quit thinking that any real or imagined flaws and discussions thereof must have something to do with what products people prefer or own. At least keep your delusions to yourselves.

The Articia S was a complete and utter failure, technologically and commercially, and so have all Terons been, and they're superceded by new products. And I suppose I'm an 'AmigaOS supporter' (Yuck! I support products by buying them, not by trolling and ignoring reality and telling others to do the same).
Nice Essen Report : Comment 93 of 115ANN.lu
Posted by itix on 05-Jun-2004 09:48 GMT
In reply to Comment 77 (ehaines):
DMA issues in A1/Linux are real and A1 mobo is the only PPC mobo in production having this problem. OS4 seems to be different matter now when they have implemented their specific solution. But there would not be need for this if Articia implemented small nice feature in HW.
Nice Essen Report : Comment 94 of 115ANN.lu
Posted by Andreas Wolf on 05-Jun-2004 09:48 GMT
In reply to Comment 90 (Sammy Nordström):
Bill Buck once published his mid to late 2002 email exchange with MAI (both CEO and CTO) about the ArticiaS. It was here on ANN IIRC. Quite an interesting read :-)
Nice Essen Report : Comment 95 of 115ANN.lu
Posted by Don Cox on 05-Jun-2004 09:52 GMT
In reply to Comment 83 (Sammy Nordström):
">The argument will be settled only when somebody finds and demonstrates a
>solution, and everyone can see how that solution works.

http://amigaworld.net/modules/news/article.php?storyid=1362"

I mean the solution in Linux, with open source code that everyone can look at.
Nice Essen Report : Comment 96 of 115ANN.lu
Posted by Don Cox on 05-Jun-2004 09:55 GMT
In reply to Comment 84 (jkdsteve2@hotmail.com):
"Why did MAI not understand what was been shown to them?"

Perhaps because they are not experts at coding Linux drivers?


"We are talking about engineers here, with logical behaviour,
analytical mindsets."

Engineers are at least as emotional as anyone else, and if you find a problem with something they have done it needs to be presented very tactfully. Not "Oi mate, this POS you made doesn't work." Nobody likes to lose face.
Nice Essen Report : Comment 97 of 115ANN.lu
Posted by Don Cox on 05-Jun-2004 10:03 GMT
In reply to Comment 91 (ehaines):
"> Everybody knows that gives trouble on the AmigaOnes.

What everyone knows, actually, is that it seems to depend on the individual AmigaOne and probably other factors like memory, etc. There are others beside me who have it enabled and it doesn't cause failures."

However, you have not run a stress test. As somebody pointed out, a day may come when some crucial file gets corrupted.

I think you are right that RAM incompatability has caused many of the problems that get attributed to the Northbridge.

Unfortunately the workings of the April chip are secret, and the AOS4 IDE drivers will be closed source (although it is known in general how the DMA is deealt with), so there are really too many unknowns.
Nice Essen Report : Comment 98 of 115ANN.lu
Posted by Sammy Nordström on 05-Jun-2004 10:12 GMT
In reply to Comment 94 (Andreas Wolf):
>Quite an interesting read

I'm sure it is, if I only knew where to look... Was it posted as a news article or as a forum post? Do you know anything about the time when it was posted that could help me locate it?
Nice Essen Report : Comment 99 of 115ANN.lu
Posted by Sammy Nordström on 05-Jun-2004 10:19 GMT
In reply to Comment 98 (Sammy Nordström):
But then, the person that posted it has made use of fabricated e-mails before, even presented it as evidence in a court of law... :-/
Nice Essen Report : Comment 100 of 115ANN.lu
Posted by Andrea Maniero on 05-Jun-2004 10:31 GMT
In reply to Comment 82 (Don Cox):
> Could be that MAI have nobody who is competent in Linux internals. After all,
> they are a chip design company. They probably didn't expect it to be a
> problem.

I usually refrain to comment on such issues. However, this statement is simply not factual. Every hardware manufacturer/designer *has* to worry about software, too. It's evident that hardware without software is useless, even more so in the field of computer components. But also the smallest fabless ICs design center has *at least* one firmwarist - if you design a component you want the buyer to be able to use/test it ASAP.
Articia was sold to perspective users in the form of the Teron board, which is next to useless without at least a working OS on it: and Linux support is, in fact, listed as a feature of the evaluation board. Moreover, buyers have to be given a complete datasheet as well as application notes, that should make the use of the product and the implementation of its features painless.
This is hardly rocket science: it's simply the way every chip manufacturer is required to work. After all, when selling something in quantities you are required to prove it's not broken. Saying something doesn't work because it uses a non standard approach is not allowed: in that case you have to give the proof it's actually working. In this case MAI should have done it by providing a fully working Linux kernel with DMA enabled.

Every chip I bought was fully documented. Every evaluation board was fully functional.

Bottom line is: you design a northbridge. You need to sell thousands of them to make it a viable design (and to pay your R&D costs). You know you'll sell them mostly as a Linux solution (when you began working on it, you were not even aware of things like AOS4). Therefore you need to have a fully working Linux ASAP, and it's not something you can save on. Otherwise the result is you'll lose contracts with manufacturers like Tratech (or Genesi, for that matters).

Kind regards,
Andrea
Anonymous, there are 115 items in your selection (but only 65 shown due to limitation) [1 - 50] [51 - 100] [101 - 115]
Back to Top