25-Apr-2024 16:33 GMT.		 UNDER CONSTRUCTION
Anonymous, there are 83 items in your selection [1 - 50] [51 - 83]
[News] Interview with Alan RedhouseANN.lu
Posted on 21-Nov-2003 15:51 GMT by TurboTrex (Edited on 2003-11-21 20:00:25 GMT by Christophe Decanini)83 comments
View flat
View list
AmigaWorld has interviewed Alan Redhouse, the managing director of the Eyetech Group, with regard to the AmigaOne motherboards. The interview includes many questions suggested by AW members.
Interview with Alan Redhouse : Comment 1 of 83ANN.lu
Message removed by Christophe Decanini for violation of ANN's posting rules.
Specific reason from moderator: flames
Interview with Alan Redhouse : Comment 2 of 83ANN.lu
Posted by Ronald on 21-Nov-2003 15:27 GMT
In reply to Comment 1 (Eva):
I still want one badly. :D
Interview with Alan Redhouse : Comment 3 of 83ANN.lu
Posted by hooligan/dcs on 21-Nov-2003 15:45 GMT
In reply to Comment 1 (Eva):
Your luck I don't own a shotgun...
Interview with Alan Redhouse : Comment 4 of 83ANN.lu
Posted by DaveP on 21-Nov-2003 15:54 GMT
In reply to Comment 1 (Eva):
Eva.

When pretending to quote from somewhere you should at least make it harder to
verify that you are a liar.

>Anyway is the discussion really serious on A****.net ?
>"PEgasos stupid owners can not use their Firewire..."

Stop making things up to cause trouble.

@Hooligan

Use it on Eva.

Dave.
Interview with Alan Redhouse : Comment 5 of 83ANN.lu
Posted by Frodon on 21-Nov-2003 16:26 GMT
In reply to Comment 1 (Eva):
Hello,

>PEgasos stupid owners can not use their Firewire...

Btw the Firewire works fine on the Pegasos, I've used it with both my iPod and a firewire camera under Linux and it worked just fine :)

Regards
Interview with Alan Redhouse : Comment 6 of 83ANN.lu
Posted by Olegil on 21-Nov-2003 16:32 GMT
In reply to Comment 5 (Frodon):
"only works with Linux" would be a good summary, then? (hint, this is what has been used as #1 reason not to buy an AmigaOne by Pegasos supporters up to now ;-) )

I'm sure the hardware works, it's just a BIT odd that noone makes MOS drivers for it, no?
Interview with Alan Redhouse : Comment 7 of 83ANN.lu
Posted by Sam Smith on 21-Nov-2003 16:46 GMT
In reply to Comment 6 (Olegil):
I suppose the MOS team are busy?

---
Sam
Interview with Alan Redhouse : Comment 8 of 83ANN.lu
Posted by Anonymous on 21-Nov-2003 16:48 GMT
Als says: "From the publicly available figures it seems that we have delivered more AmigaOne boards than any other 'open' [...] standard PC form factor board supplier"

I wonder why he makes the fine destinction with form factors every time he talks about sales. As everybody knows, the Pegasos is NOT standard form factor but micro-atx. So is he talking about Pegasos or not?! He usually implies he is (in his remarks on agp speed and sniggering at free boards) but these statements are so carefully worded. If he does make such fine distinctions, he truly is a cunning fox ;-) I still wonder how many A1 were sold. Any idea how many different people post on the A1 mailing list?
Interview with Alan Redhouse : Comment 9 of 83ANN.lu
Posted by Sam Smith on 21-Nov-2003 16:50 GMT
In reply to Comment 1 (Eva):
This is just ultra sad. Why post this? What is wrong with formulating an intelligent comment? This sort of thing just devalues anything with your name attached.

How about an apology and a return to some credible comment?

---
Sam
Interview with Alan Redhouse : Comment 10 of 83ANN.lu
Posted by Frodon on 21-Nov-2003 16:50 GMT
In reply to Comment 6 (Olegil):
Hello,

@Olegil
> "only works with Linux" would be a good summary, then? (hint, this is what has been used as #1 reason not to buy an AmigaOne by Pegasos supporters up to now ;-) )

Well at least if you really don't want to use Linux you can :)
But personnally I use Linux to complete what I can't do with MorphOS, exactly the same as I did with my A1200, I used LinuxPPC to complete what I can't do with AmigaOS.

But in the case of FW, it was more a test, I usually prefer to use my Mac for my iPod and to use a FW video cameras because iPod is a lot better supported (particularly to use its MP3, calendar and address book features) and digital video softwares are a lot better than what you can find in Linux or even Windows :)

Regards
Interview with Alan Redhouse : Comment 11 of 83ANN.lu
Posted by Christophe Decanini on 21-Nov-2003 16:53 GMT
In reply to Comment 6 (Olegil):
"I'm sure the hardware works, it's just a BIT odd that noone makes MOS drivers for it, no?"

Implementing a Firewire stack takes time and money.
If Eyetech and Genesi want something more than a Firewire mass storage stack they will have to sell a lot of boards.
We will see who come up first with a solution and how good the solution is.
Interview with Alan Redhouse : Comment 12 of 83ANN.lu
Posted by Christophe Decanini on 21-Nov-2003 17:11 GMT
In reply to Comment 8 (Anonymous):
"IDE DMA drivers have been modified to make them behave properly on the AmigaOne and a universally applicable OS patch is now in test."

What is surprising me the most is that at this time the infamous Via/Articia feature/bug is still not used/fixed and that hundreds to thousands AmigaOne owners do corrupt their data.
I heard that the Amigaone sold at Alchimie 3 where still failing the MD5SUM test(confimed in this interview as Alan say there is a patch in test).

Either these customers do not trigger the bug very often (maybe just using their machines to do things involving small data IO such as browsing, IRC, etc) or they don't know /do not care about keeping their data safe.

"We believe that this is a much better solution than to throttle back the simultaneous memory access with a hardware dongle as has been tried by others."

I do not agree as I have used a Peg1 + April 2 safely since april and will certainly have a Peg2 when the new patch will be released for the AmigaOne.

Hopefully this time the upcoming patch will definitely fix the problem as we where told it was fixed months ago with Bios/Kernel/Drivers updates.
Interview with Alan Redhouse : Comment 13 of 83ANN.lu
Posted by Sam Smith on 21-Nov-2003 17:24 GMT
In reply to Comment 12 (Christophe Decanini):
I think when Alan was talking (indirectly ;) about the April chip he wasn't suggesting that 'throttling' it made it less reliable with data but simply that it made it slower than doing full simultaneous transfer.

Now I do not profess to know any of the specifics but from what I gleaned in the interview it would seem to suggest that the April solution works but it is working around the problem rather than solving it. I don't own either board so I cannot verify this for myself.

Is this correct?

---
Sam
Interview with Alan Redhouse : Comment 14 of 83ANN.lu
Posted by SlimJim on 21-Nov-2003 17:35 GMT
@ those that feel targeted

Some among you might prefer another hardware+OS solution than this, but at least
respect those that *do* prefer this by not turning this into a flamewar. And
those of you supporting this hardware, please avoid bloody *baiting* the other
side, will you? This crap happens EVERY time :-( ...!

Personally, I find this a very good interview. Alan is making sure to be precise
and as diplomatic as possible in his replies. He sounds very professional, in my
opinion (<-note!).
.
SlimJim
Interview with Alan Redhouse : Comment 15 of 83ANN.lu
Posted by Christophe Decanini on 21-Nov-2003 17:42 GMT
In reply to Comment 13 (Sam Smith):
The April does correct the problem shown by the MD5SUM test.
I don't know how the April does effect performance.
My HD can get as far as 40 MB/s, the mem tests have been published.

What do Amigaone owners do on their ends ? Do they disable DMA for the HD ?
In this case the performance penalty is definitely higher. If not the reliability may be affected.

When BBRV said that the Pegasos was performing better than the AmigaOne all the people (red and blue) rushed on the benchmarks while "performing better" may have been for performing with more reliability.
Interview with Alan Redhouse : Comment 16 of 83ANN.lu
Posted by Anonymous on 21-Nov-2003 17:43 GMT
In reply to Comment 13 (Sam Smith):
> it would seem to suggest that the April solution works but it is working around the problem rather than solving it.

What's that, advanced logic? "Working around the problem" vs "solving it": Is there a significant difference? There is a bug in Articia, so strictly speaking you can not solve it unless you take the chip apart. But apparently you can work around it outside Articia. The problem with Articia apparently is that the DMA logic is not always synced with the cache consistency logic in the cpu, ie. the data in the cache does not consistently represent the data in the main memory if DMA and CPU concurrently access memory. That's Bad (TM) or, in Eyetech speach, an Amazing Feature (TM). If the system was working as it is supposed to work, the DMA logic would invalidate cache content. I assume that the April fix snoops the address bus and detects DMA and then does something about it, like flushing the cache before the DMA is permitted to proceed!? I have to admit I have no clue re hardware and CPU :-)
Interview with Alan Redhouse : Comment 17 of 83ANN.lu
Posted by Sigbjørn Skjæret on 21-Nov-2003 17:48 GMT
In reply to Comment 13 (Sam Smith):
Well, either way, why would you rely on someone who's not only a competitor, but also seems to consistently try to explain away a bug in hardware by claiming it happens because the drivers are *not* using a supposed "feature" of said hardware (a feature should *add* functionality, not *detract*, and if not using the "feature" means the regular *documented* way doesn't work properly, then it's a *bug*, no matter what you say to try to explain it away) to explain reliably to you how someone else's fix works?

IMO these snide remarks are pretty pathetic, especially given the fact that he can't even admit an obvious bug, and even when trying to cover it up does such a laughable job that most upwardly mobile people see right through it...

Oh, and I see that they weren't able to fit all the things they claimed would fit (Firewire, USB 2.0, RAID and whatnot) on the mini-itx board, but instead opted for maybe making an add-on daughterboard with these features later on .. wonder why so many were so hostile when it was suggested this wouldn't fit on their already crowded board when it was first announced... :P


- CISC
Interview with Alan Redhouse : Comment 18 of 83ANN.lu
Posted by Christophe Decanini on 21-Nov-2003 17:50 GMT
In reply to Comment 14 (SlimJim):
"Personally, I find this a very good interview. Alan is making sure to be precise and as diplomatic as possible in his replies. He sounds very professional, in my opinion (<-note!). "

I think he could avoid all the Genesi alusions. He is just helping to have more flames and reaction such as Eva's.
What is the point to comparing the AmigaOne to the Pegasos 1 that has been sold out long time ago ?
I am also sure that the part on the feature that makes an AmigaOne as fast as a 3 times more clocked PC will make some people react too.
Interview with Alan Redhouse : Comment 19 of 83ANN.lu
Posted by Christophe Decanini on 21-Nov-2003 18:18 GMT
After reading this interview for a second time I wonder how much will be the miniitx AmigaOne with OS4.
The current Amigaone are far from cheap and they are a special offer (Once OS4 is released you will have to pay for AmigaOne + OS4).
Hopefully they will get a decent price so that a full configuration can be in a reasonable range.
Interview with Alan Redhouse : Comment 20 of 83ANN.lu
Posted by SlimJim on 21-Nov-2003 18:38 GMT
In reply to Comment 18 (Christophe Decanini):
Well, perhaps Eva is not a very good measure on if an item is flame-instilling or
not ... Just posting a plain text file containing one of the words "AInc",
"Eyetech", "Articia" or "Hyperion" would still cause him/her to start making
bizarre, ill-researched accusations and innuendos.

(I know, that's flaming. Sorry. But I do get kind of tired of that Eva character
from time to time)
.
SlimJim
Interview with Alan Redhouse : Comment 21 of 83ANN.lu
Posted by Christophe Decanini on 21-Nov-2003 19:20 GMT
In reply to Comment 20 (SlimJim):
Well, thankfully not everyone comments like Eva.
In my case if the interview would have less exposed how Hyperion / Eyetech are better at implementing / fixing bugs than Genesi I may have had a better perspective on their plans instead of arguing on how the spoke about genesi.

We have already been told many times that there were no bug or that the bug were fixed in software.

First on Sept 30 2002 Ben Hermans said "For the record, there is no Articia bug.
That's what you get when you buy in your firmware and don't actually know how your Northbridge and Southbridge work."

Then a week later Eyetech had the wire fix and said proudly that the Pegasos does not have such (G Carda fixed this problem with MAI, April was in the making).
Then Genesi said that there were still some bugs while Eyetech/Hyperion denied.
Then Eyetech got the new release of the articia (the supposed bug free one).
Then the MD5Sum test was exposed; AmigaOne developers acknowledged the bug and told some time later that they had a fix.
Then BBRV told about another way to expose the bug then Hyperion Eyetech said that it is a feature that they will address.
Now Eyetech is telling that with using such feature the AmigaOne will be better than using an "hardware dongle" (read April).

Undoubtfully we can say that there are some serious problems and we understand why Genesi dropped this northbridge. The people screaming FUD everytime a problem has been exposed should have better calmed down.
Interview with Alan Redhouse : Comment 22 of 83ANN.lu
Posted by Anonymous on 21-Nov-2003 19:28 GMT
"an optical mouse has no balls"
Interview with Alan Redhouse : Comment 23 of 83ANN.lu
Posted by Anonymous on 21-Nov-2003 19:58 GMT
In reply to Comment 22 (Anonymous):
Neither does my non optical mouse, it's just got one.

...

Whada ya mean killer of all fun!!!

;-)
Interview with Alan Redhouse : Comment 24 of 83ANN.lu
Posted by Sam Smith on 21-Nov-2003 20:12 GMT
In reply to Comment 15 (Christophe Decanini):
Thanks for clearing that up! :)

---
Sam
Interview with Alan Redhouse : Comment 25 of 83ANN.lu
Posted by Sam Smith on 21-Nov-2003 20:20 GMT
In reply to Comment 16 (Anonymous):
Yes. I believe there is a clear difference between working around a problem and solving it. 'Working around' a problem means that the problem is still there and is not an optimal solution. Solving it means that it isn't there and everything is working optimally. In this case the workaround is to 'throttle' the data thereby making the system slower than what it was originally intended to be. This is a 'work around' and not a solution. The solution would be to have a totally fixed chipset. The problem isn't with Eyetech or Genesi as they are just getting around the issue as best as they can.

---
Sam
Interview with Alan Redhouse : Comment 26 of 83ANN.lu
Posted by Sam Smith on 21-Nov-2003 20:30 GMT
In reply to Comment 21 (Christophe Decanini):
You are right in that it doesn't help if people turn this into 'mines better than yours' argument when we all as users simply want to know the facts about the issue.

I doubt that either solution will have any serious problems. We know already that MOS is stable on the Pegasos. I suppose we have to accept that both systems are in their early stages and a few teething problems like this are to be expected.

It is good to see the level of flames dying down when these issues come up so that they can be properly discussed to everyone's benefit. :)

---
Sam
Interview with Alan Redhouse : Comment 27 of 83ANN.lu
Posted by Christophe Decanini on 21-Nov-2003 21:34 GMT
In reply to Comment 26 (Sam Smith):
" It is good to see the level of flames dying down when these issues come up so that they can be properly discussed to everyone's benefit. :)"

Well, it is friday night so i guess people heve better things to do than hanging here ;)
Interview with Alan Redhouse : Comment 28 of 83ANN.lu
Posted by Sam Smith on 21-Nov-2003 21:49 GMT
In reply to Comment 27 (Christophe Decanini):
We can't possibly go out as we've got to get an early night in ready for the final tomorrow! :)

---
Sam
Interview with Alan Redhouse : Comment 29 of 83ANN.lu
Posted by JoannaK on 21-Nov-2003 23:00 GMT
In reply to Comment 18 (Christophe Decanini):
CD: AlanR can't compare AmigaOne to Pegasos II cause it would only show how much they are behind in technology. I know that some will (obviously) point out that Pegasos II has not been delivered to masses yet.. but I'm sure it'll appear to end users months before Eyetechs 'new' lite and mini boards. (Commenting on OS woudl be off topic anyhow).

And for his technology knowledge.. Well. Lets just be polite and admit he's a lot better on salesman than hw/sw engineer. .-)
Interview with Alan Redhouse : Comment 30 of 83ANN.lu
Posted by ikir on 21-Nov-2003 23:23 GMT
Very nice interview! :D
Interview with Alan Redhouse : Comment 31 of 83ANN.lu
Posted by Darth_X on 22-Nov-2003 02:04 GMT
In reply to Comment 2 (Ronald):
@Ronald (24.202.65.121)

>> I still want one badly. :D

AmigaOne, Pegasos II, A1-lite? Or all of them? ;)

ps: you're on the east coast of Canada, right?
Interview with Alan Redhouse : Comment 32 of 83ANN.lu
Posted by Darth_X on 22-Nov-2003 02:21 GMT
"2) How many AmigaOne boards have been sold so far and what is your opinion on the demand for the current boards and upcoming Micro1A boards?

Alan: From the publicly available figures it seems that we have delivered more AmigaOne boards than any other 'open' (ie not IBM or Apple) PPC-based standard PC form factor board supplier. And in terms of real sales for real money (which in my view is the ultimate measure of success - anyone can give boards away) we really seem to have no significant competition to date.

The MicroA1 is aimed at a different marketplace entirely. It is - at the same time :

- a potential (ie first it needs a stable OS4 + applications) entry level home computer and (initially retro-ish) games console
- a modular embedded system/industrial controller board (with a variety of I/O and cpu options) running embedded OS4 or Linux from flash ROM
- a very low power consumption, low cost of ownership thin client (running under OS4 or Linux as appropriate) for Windows or Linux application servers. The client software for Linux has already been written and will be ported to OS4 (allowing the same delivered performance for lower CPU specification/cost) in due course.

Ultimately I believe that this market will develop to see the MicroA1 technology licenced to the big far-eastern LCD screen manufacturers to allow them to incorporate thin client (and OS4 capability) directly into their own products.
"

Some good points by Alan, but he's still not telling us the sales numbers for the AmigaOne. Could someone post these 'publicly available figures' I honestly don't understand the purpose of hiding the actual sales numbers. No excuses please!
Interview with Alan Redhouse : Comment 33 of 83ANN.lu
Posted by Ronald on 22-Nov-2003 02:47 GMT
In reply to Comment 31 (Darth_X):
@Darth_X (24.108.69.230)
<<<AmigaOne, Pegasos II, A1-lite? Or all of them? ;)

ps: you're on the east coast of Canada, right?>>>

I'm interested in the AmigaOne Lite only. Mainly to view movies from the living room.
Interview with Alan Redhouse : Comment 34 of 83ANN.lu
Posted by Darth_X on 22-Nov-2003 03:02 GMT
In reply to Comment 33 (Ronald):
>> I'm interested in the AmigaOne Lite only. Mainly to view movies from the living room.

Sounds like a good idea. ;)
Interview with Alan Redhouse : Comment 35 of 83ANN.lu
Posted by Bodie_CI5 on 22-Nov-2003 05:46 GMT
In reply to Comment 34 (Darth_X):
As well as for the bedroom TV.
Interview with Alan Redhouse : Comment 36 of 83ANN.lu
Posted by Anonymous on 22-Nov-2003 06:14 GMT
In reply to Comment 29 (JoannaK):
How about we also remember that we haven't heard a *single* report of a Peg2 in beta testing? knowing how BB operates, if there was even one person working on a Peg2 (porting Linux, or BSD, or even MOS) he'd be shouting it from every rooftop. The silence is deafening.

And Alan *is* a salesman, I don't think he ever claimed to be an engineer.

Why does every single thread on this site have to be Genesi vs the world? the very first post was an attack, and it went downhill from there. Do some people feel threatened that they're not the only game in town? (and will someone please feed Eva her medications?)
Interview with Alan Redhouse : Comment 37 of 83ANN.lu
Posted by DaveP on 22-Nov-2003 07:56 GMT
In reply to Comment 36 (Anonymous):
I guess some people are feeling threatened and off balance, which I guess is
going to be the case until the Pegasos2 ships to end users.

Nothing we haven't seen before, I reckon its yet to reach a fever pitch.
Interview with Alan Redhouse : Comment 38 of 83ANN.lu
Posted by IanS on 22-Nov-2003 08:20 GMT
In reply to Comment 29 (JoannaK):
>>cause it would only show how much they are behind in technology

Care to clarify JoannaK?
Interview with Alan Redhouse : Comment 39 of 83ANN.lu
Posted by ikir_spirit on 22-Nov-2003 08:33 GMT
>>cause it would only show how much they are behind in technology

Care to clarify JoannaK?
--------------

She can't..... she's only trolling as usual :-/

Again: GREAT INTERVIEW, don't feed the trolls.
Interview with Alan Redhouse : Comment 40 of 83ANN.lu
Posted by Don Cox on 22-Nov-2003 08:47 GMT
In reply to Comment 25 (Sam Smith):
"The solution would be to have a totally fixed chipset. The problem isn't with Eyetech or Genesi as they are just getting around the issue as best as they can."

I think you are missing Alan's point. He is saying that there is _no_ bug in the Articia chip, but there is a design feature which is absent from the chips used on typical x86 boards, and is not properly supported by most Linux drivers.

In other words, the bugs are in Linux, not in the chip. That is what Alan is saying. If you disable that design feature, then the Linux bugs will not appear - that is the workaround. Fixing the code is the real fix.
Interview with Alan Redhouse : Comment 41 of 83ANN.lu
Posted by Don Cox on 22-Nov-2003 08:49 GMT
In reply to Comment 29 (JoannaK):
"And for his technology knowledge.. Well. Lets just be polite and admit he's a lot better on salesman than hw/sw engineer. .-)"

Alan does have an engineering background, unlike Bill and fleecy.
Interview with Alan Redhouse : Comment 42 of 83ANN.lu
Posted by Don Cox on 22-Nov-2003 08:53 GMT
In reply to Comment 32 (Darth_X):
"Some good points by Alan, but he's still not telling us the sales numbers for the AmigaOne. Could someone post these 'publicly available figures' I honestly don't understand the purpose of hiding the actual sales numbers. No excuses please!"

The "publically available figures" are those for the Pegasos. He is saying he has sold more than the published sales of the Pegasos. He isn't saying how many more.
Interview with Alan Redhouse : Comment 43 of 83ANN.lu
Posted by IanS on 22-Nov-2003 08:54 GMT
In reply to Comment 40 (Don Cox):
Spot on Don.
Interview with Alan Redhouse : Comment 44 of 83ANN.lu
Posted by Kronos on 22-Nov-2003 09:10 GMT
In reply to Comment 40 (Don Cox):
A feature is normally documented at the time a chips released (or even before), not
after 2 years of denying and threatening HW-developers trying to fix it.

A feature is normally something that does no harm when it is not used.

A feature is normally something that does bring some sort of benefit.

Now compare an A1 with a late SDR-based Mac, and see which performs better
(under Linux), and yes the Mac does have DMA, but it won't need special drivers for
an obscure "feature". And remember that SDR-based (desktop) Macs are long gone
from the production lines.

The "feature" is that the Articia doesn't (reliable) inform the CPU about DMA-transfers
and the esulting need to update cache. The SW-solution is nothing more than telling
it to the CPU by hand, a method that will hardly be better/faster than having it implemented
into the HW.

The Articia is 3 year old tech which is now just about to become useable to it's full spec
(DMA has been on the feature-list from day 1).

HOORAY !
Interview with Alan Redhouse : Comment 45 of 83ANN.lu
Posted by Don Cox on 22-Nov-2003 09:30 GMT
In reply to Comment 44 (Kronos):
"A feature is normally documented at the time a chips released (or even before), not
after 2 years of denying and threatening HW-developers trying to fix it."

I don't see why hardware developers would be trying to "fix" it at all. The problem seems to be bugs in Linux.



"A feature is normally something that does bring some sort of benefit."

I would have thought that simultaneous and transparent memory access by the CPU and a PCI bus master would speed things up?
Interview with Alan Redhouse : Comment 46 of 83ANN.lu
Posted by Kronos on 22-Nov-2003 09:37 GMT
In reply to Comment 45 (Don Cox):
@Don

Don't know if you did see Ron Heinleins comments in the last of these threads.
He basiclly stated that he had discovered this behaviour, and recommended a
workaround. MAIs answer wasn't "it's a feature", or "we will fix it", but threats for
the case he would release this workaround.

a) There can't be real simultan RAM-access, aslong you don't have special 2 channel
chips.
b) Pseudo-simultan RAM-access is quite common even in the x86-world, but it only makes
sense when the chipset overs enough bandwith AND if caches are taken care of.
Interview with Alan Redhouse : Comment 47 of 83ANN.lu
Posted by itix on 22-Nov-2003 10:26 GMT
In reply to Comment 40 (Don Cox):
> In other words, the bugs are in Linux, not in the chip. Yeah, sure.
Interview with Alan Redhouse : Comment 48 of 83ANN.lu
Posted by Don Cox on 22-Nov-2003 10:58 GMT
In reply to Comment 46 (Kronos):
It does seem that Mai have been remiss in not employing two or three coders to port Linux and make sure everything worked correctly. That job should not have been left to customers like Genesi or Eyetech.

After all, there was no other likely OS for such hardware other than Linux.

MAI seem to have gone for a fairly adventurous design, and then not tested it quite thoroughly with a real life OS. OTOH it isn't surprising if it takes a new company a year or two to sort out all the problems presented by a new product.

I think we now have to wait and see if the software bug fixes that Alan mentions do solve the problems. That will probably require some quite tedious testing by somebody well familiar with Linux.
Interview with Alan Redhouse : Comment 49 of 83ANN.lu
Posted by itix on 22-Nov-2003 11:23 GMT
In reply to Comment 48 (Don Cox):
> I think we now have to wait and see if the software bug fixes that Alan mentions do solve the problems.> That will probably require some quite tedious testing by somebody well familiar with Linux. We will see. If software fix doesnt work you cant blame Linux anymore. Mainly problem in testing this fix is that for most users DMA corruption doesnt occur very often. This software fix looks interesting anyway.
Interview with Alan Redhouse : Comment 50 of 83ANN.lu
Posted by Anonymous on 22-Nov-2003 12:35 GMT
"anyone can give boards away"

No they can't! :-)
(Wonder if Eyetech could do it, even if they wanted to ...?)

Alan is comparing Eyetech to Genesi, and it's clear that Eyetech sees this fan-community-market as the end goal where they will make their money, while Genesi sees this market as a furtile grow ground where they can seed development for the *real* markets. That explains the difference in a 800€ price tag contra "Phree" boards ...

"We are currently looking at the feasibility of adding the 'enhanced I/O' (RAID, gigabit ethernet, USB2, IEEE 1394 etc) on a separate plug-in daughter board."

Oh, so then this "prototype" that has been shown, with the **Articia S**, might actually be the final end user board after all? How on earth could a Articia S coupe with RAID, gigabit ethernet, USB2 and IEEE 1394 ??!?! These buzz words may look nice in ads and marketing but it will not live up to it's promised specifications. And they will all live on the same PCI card? Wow! Wonder if that daughter card will also have a fast northbridge (like the Marvell), with high bandwidth to some fast memory, and it's own CPU ...? :-P

"The Articia S northbridge used in the AmigaOne supports 2x AGP speeds."

Not all the time/for all actions. Sometimes only 1x speed ...

"For example, the actual delivered speed difference between, say, 2x and 4x AGP is generally accepted to be only around 15%."

Huh? If you do some real life tests with a modern graphic card with 128 MB memory you will find that the real life difference between AGP 1x and APG 8x is only a few percents in most games, even the newest and most powerful PC 3D games, and this is nothing a 3D player will notice when playing their games. I have tested this myself! AGP speed numbers are only a marketing hype that the regular clueless PC buyer might fall for. Hardware support and drivers for high performance modern graphic cards (and games that makes use of their features) is more important.

"There have been two problems with DMA, neither of which is down to any bug in the Articia 'S' northbridge."

I have a hunch that they will be prooven wrong on this point in some time. But perhaps not, they might find new escape goats, like "it's not a bug, it's a feature", and "this is only a driver issue" (driver=patch to disable/work around the hardware bugs) in the same "honest" spirit they have shown their customers up to this point.

"Of course we are also looking at several possible design specifications for future products on an ongoing basis"

We = MAI

"A1 specs said that the ROM would be non reflashable (in fact an PROM)"

Insane, by todays standards ...!

"Prices will be announced when the products are ready to go into production."

If the price for a non-gigabit, non-raid, non-usb2, "non-etc" motherboard with the Articia S is above €200, then don't even bother. Poor specifications and poor expansion possibilities will make it difficult. The prices on mini ITX are low for these reasons, and €200 is *not* a low price. With the higher specifications, and with a northbridge that can handle it, you could perhaps charge even slightly more than a Pegasos 2, despite the fact that the customer won't be able to expand it in the future.
Anonymous, there are 83 items in your selection [1 - 50] [51 - 83]
Back to Top