Pegasos to go on sale Monday / Pegasos II to ship September 2003

[System]

My thoughts on this are that no product should be advertised as "obsolete" or "soon to be replaced" for any reason. It's good that they are looking forward to the next design, but that should be transparent to the end user. When the PegasosII is nearing a presentable stage, then it should have been announced that the original Pegasos would be phased out

And then a Zillion of retarded users will start to call BBRV a liar.

[downix]

For instance I know that the PegasosG3 performs similar to the AmigaOSG3-SE (Linux). So I have my severe doubts about i.e. a G4 AmigaOne-XE or Teron board performing worse than a G3 one.

And you've done labratory-condition benchmarks to proove this?  If so, please, post the results here for everyone to see.  Please make sure which revision of the motherboard, Rev1.0, Rev1.0A or Rev1.0B you used for the test, what peripheral cards are included, which OS and most of all which industry-approved benchmarks you used for the test.  I would very much appreciate having such a test done, as would many others.

[anarchic_teapot]

Wait. Are you telling me that you had ten months the
machine and did not find the first bugs which then
later need to be fixed in the A1 SE and neither found
the bugs that have been fixed in the new Articia S
revision? Is that what you*re telling? 10 months?
Wow.

No. The first bugs were so arcane (ie did not affect the use of the mobo as an ordinary desktop 'puter) that we didn't notice them in all of 5 months' use. Gerard Carda did, bully for him (no, seriously, he did a good job).
As for the new revision - which comprises the fixes for the bugs in question (2 of them) - I would be interested to learn in what way it is buggy, especially as only A1XE owners have one at present.

As for the "other bugs" - so noisily trumpeted as being fixed [1], yet MAI,  who readily accepted the other 2, claim to be unable to reproduce and therefore dispute their existence - are they really a problem with the Articia S, or a problem with the design of the Pegasos (or its firmware)?

I'm still waiting for proof, not innuendo, FUD, or any other form of denigration, of the Articia S being a disaster, bearing in mind I use the damn thing on a daily basis and find it works rather well. As do quite a few other people, BTW.

[1] though if the wonderful April chip does fix these problems, why can't the Pegasos 1 use a G4 as originally promised ? Of course, this may have nothing to do with the hardware, but then the obvious reason would then be that Genesi can't afford to pay for the chips, having spent all its money in marketing. Not a good sign either.

[Troels_E]

@ Senex

"Well, regarding Genesi's latest statement it's not a matter of not getting them, but of not wanting them because of further bugs..."

Don't tell me you actually believe that?  

Come on, at least agree that it sounds like a very bad excuse.

I find it hard to believe that Genesi stops the production on their own free will, after selling only 1500 pegasos'es.
If the MAI bridge is so bad why produce the last batch of PEGS?

I find it much more likely that MAI simply said "no more Articias for you". They could have several reasons to do this.

[System]

@ Downix

And you've done labratory-condition benchmarks to proove this?

LOL, Downix do you really think you would need labratory-condition benchmarks? I would like to use my AmigaOne within my spare bedroom instead.  :-D

With regard to the board model used, he must have compared the Pegasos to the latest Teron models. I know that Genesi owns an old evaluation board with a Softex ROM. You don't possibly mean to say that they have used this old board for their "labratory-condition" benchmarking?

If so, please, post the results here for everyone to see.

Why don't you ask Genesi  to back up their wild claims?  LOL

[JoannaK]

@MikeB

Personally I highly doubt this. The executives of Terra Soft, Eyetech and Mai were not at all happy with Bill's FUD approach with regard to their efforts.

Well.. how come Genesi still got those 400 Articia-S? If there had been such skism as you suggest, why Mai bothered selling and sending them to Genesi at all? They cound have used those chips as a landfill  and concentrate to those bigsellers like Eyetech  :-P

For the rest of your post.. I'll think keep on commenting for now...   Hold that thought for couple weeks.

[Troels_E]

Well.. how come Genesi still got those 400 Articia-S?

Could be the last shipment according to contract?

Or maybe MAI knew that they wouldn't have other costumors for it as Eyetech wanted to wait for the new fixed ones.

There could be a zillion reasonS, all we are doing here is just wild speculation:-D

[downix]

@ anarchic_teapot

A disaster for one company is business as usual for another.  Genesi might have specific needs from their northbridge that Mai promiced but cannot deliver.  (The RAM issue is still one that looms, for example)  Genesi claimed that Mai did not work as advertised.  This does not preclude the Mai from working, just not as-advertised.  Where Mai is deficient may not appear in a desktop solution, for example.  But remember, Genesi needs more out of their chipset vendor than *just* the Pegasos, as shown by the Eclipsis and STB deals.  If Mai cannot deliver as-promised, then Genesi has wisely chosen a vendor which can deliver what was promised.  

As for the G4 question, they never said it *couldn't* run a G4, but that Genesi was not going to deliver a G4 card for it.  Commodore never made 68060 nor PowerPC cards for the Amiga either, yet I once owned a card with both of these processors on it at one time for my A1200.  Also, as noted in the list, those owners of a Peg I can upgrade to a G4, by trading in their Peg I for a Peg II and paying the $200 for the G4 card (which was the original price anyways).

[System]

@ JoannaK

f there had been such skism as you suggest, why Mai bothered selling and sending them to Genesi at all?

As can be gathered from earlier message board postings with regard to these older articia chips, according the Bill Buck, Mai and their partners would not have much use for these older chips, as all new AmigaOne-XE and Teron boards will be shipped with newer Articia chips anyway.

[downix]

Why don't you ask Genesi to back up their wild claims?

You're the one challenging it.  

If you insist on a benchmark comparison, however, I can ask around to find a 3rd party willing to make one, if you'd back me up on it being a fair and unbiased test.  I'd also need someone with an A1 board to compare with, preferably a few different variations on such (the original G3 model, single G4 model, dual G4 model) and I can discuss with the other pegasos owners to get a comparison of the now 3 revisions of the board.  It's the only way to be fair, agreed?

[hnl_dk]

@downix

Why do you say that it is MikeB who is challenging it???

You're the one challenging it.

It was bbrv who was saying that the Pegasos I (G3 - oh yes, no G4 ) is faster than the MAI/Eyetech G4 ... he can not say that without any tests ... that is a false claim ... why should MikeB prove that the AmigaOne board is the fastest one, when it is bbrv that has gone wild in the darkness?

[Skyraker]

I still speculate that MAI has pulled the plug on Genesi, mostly because of the eyetech contract.... If i were eyetech I would lean on MAI...

I seriously doubt  genesi has discounted these chips until late this year because theyre not good enough.. i think they've had no choice in the matter.... whatever it's bad news either way....

One platform *must* succeed and more delays are only prohibitive....

To stop production until late 2003 (read 2004) is suicide on any other grounds than unavailability..... therefore i read that Genesi can not source these chips... which is a shame.. it a nice piece of h/w

We're all in the same boat...

[zacman]

>yet MAI, who readily accepted the other 2, claim to
>be unable to reproduce

So you don't remember that MAI denied the first bug
too? Only after a long period of time they said that
they can reproduce it and also only because Mr
Carda was in Fremont, CA for two weeks...

>or a problem with the design of the Pegasos (or its
>firmware)

Like the first bug Mr Carda found?

>I use the damn thing on a daily basis and find it
>works rather well.

There are other people also using the Pegasos and
it works fine for them. But that doesn't mean you
don't actually have to fix the bugs. Oops this ram
module doesn't work? Well it's not a bug - it's a
feature, you know low power ...

>why can't the Pegasos 1 use a G4 as originally
>promised ?

They could. But it doesn't make sense just to give
them a CPU card so that people can feel good when
you know that there could be much more speed if
there was another NB. It would be unfair to all of
those who want the real speed of the G4 CPU.

[zacman]

>Could be the last shipment according to contract?

Ehm the shipment of these Articia chips has been
worked out together with MAI at CES 2003.

I'm sure if there were chips available in numbers
(read more than 25/50 or 100) and be bugfree then
Genesi would take.

[System]

@ Downix

I can ask around to find a 3rd party willing to make one, if you'd back me up on it being a fair and unbiased test.

Be my guest. These tests should be well documented and easily reproducable and I would happily accept them as being genuine.

You're the one challenging it.

So you think I should blindly accept everything BB feeds us? As I said, based on my personal experiences, the performance of currently available G3 solutions are about equivalent.

With regard to processor performance tests, I rather believe the benchmarks provided by IBM and Motorola. Do you truly understand what it would take to objectively test all possible benchmark scenarios? For instance did these "tests", BB refers to, include Altivec optimised software tests? Altivec optimisations can result into performance gains up to 60%, but the performance gains do differ for different types of software.

[Herewegoagain]

They could. But it doesn't make sense just to give them a CPU card so that people can feel good when you know that there could be much more speed if there was another NB. It would be unfair to all of those who want the real speed of the G4 CPU.

LOL!  It's so funny to hear all of this now.  Before this announcement the Pegasos was "all that, and a bag of chips".  Now it's dumping the current chipset because it's so buggy they cannot get the real speed of the G4 CPU!!!!  Please stop trying to make excuses for Genesi and Pegasos.  It's not the worst thing that they are redesigning it.  But please at least be willing to look at all of the companies involved Eyetech, Mai, IBM, Terrasoft, etc... and they are not having these problems that the Pegasos is having.  Just for once admit that it could inherently be a design flaw in the original Pegasos board.  It wouldn't be a crime, and certainly not the first time this sort of thing has happened.  

People on "both sides", please listen.  These boards are new and in the greater scheme of things, RELATIVELY UNTESTED in everyday use.  That's not to say completely untested, but problems will continue to arrise from both side as the user base grows.  The orignal Amiga's had issues that had to be worked on as well.  Genesi obviously see's a need to go back and redesign.  Good.  While they are at it, they will make improvements based upon what is currently available.  That makes sense.  This will likely happen in the next revision of AmigaOnes as well (it's already happened to a lesser extent).  Let's just see what the next few months hold for us all.  

This is good news in the long run.  As it ups the specs that AmigaOne systems will have to match.  Maybe the next version will use DDR and AGP4-8X and etc.  By this, Genesi are pushing Eyetech to start on the next big thing (if they haven't already).  That can only be good for us all.