@AMiGR

I never said I'd be willing to bet on them delivering this year, I just refuted your statement that increasing board size to 17x19 cm would mean a major redesign.

PCBs for proto series can be manufactured overnight, hundreds might be doable in less than a week. There's a complete lack of any facts in any of your speculations.

1+1 = 2 (for most purposes), but you still gotta make certain it's 1 and 1 you're adding, just jumping to the conclusion "2" without checking the integers on the left hand side will very likely not produce a valid outcome. So do try to hold back your conclusions until the facts are known.

It actually IS possible to get a design right the first time if you follow all the chip specifications and application notes, allthough I must confess it's rarely done

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@TheDungeonDelver

Naahh keep it cheap so more people can afford it.

_________________
AmigaOS or MorphOS on x86 would sell orders of magnitude more than the current,
hardware-intensive solutions. And they'd go faster.-- D.Haynie

@olegil

No. Unless you have a FINAL prototype now, there's no chance that you can release a *USER* board before the end of the year and still have it tested enough... All the problems the early AmigaOnes had were due to the lack of PROPER testing, another such hit and this market is *GONE*. *THAT'S* my point. You can produce early prototype PCBs overnight, you cannot prototype something in October and release it TO USERS in December. I never said that changing the size of the PCB itself was a problem... I said that doing it now, so late, without having proved the original design in the first place, is stupid.

BTW, there has been another change in the specs anyway... They are adding an AGP slot and changing more than the size of the PCB. Now, tell me that a December release is still even remotely realistic...

Last edited by AMiGR on 18-Oct-2005 at 10:35 PM.
Last edited by AMiGR on 18-Oct-2005 at 10:34 PM.

_________________

@AMiGR

Time will tell

_________________
Site admins are people too..pooff!

G4 would be nice.

Still a case for wait and see for me such as price, size, date and bugs (if any).

This AGP port in PCI mode thing. Can all AGP cards that will plug into this (AGP version 1?) port run in PCI mode, like it is some sort of standard, or is it just a few that do PCI mode?

Last edited by Intuitioned on 18-Oct-2005 at 11:50 PM.

_________________

@Intuitioned

Yes, AGP mode is something that you have to enable. I don't know about cutting-edge GFX boards, but more common stuff comes up in PCI mode first, then the SW has to tell it what AGP speed to use, etc.

_________________
cheers
tony

Hyperion Support Forum: http://forum.hyperion-entertainment.biz/index.php

@tomazkid

It will and as I've already said, if they prove me wrong, I will apologise, in public.

_________________

@olegil

Let's be realistic, though, by the time a small batch of prototypes is built, farmed out to testers with Linux and updated to fix the inevitable bugs, then a production run made, your average AW reader isn't going to see one this year.

A HAL still has to be written before it will run OS4, and only when OS4 is running on it will we be able to find the nastier bugs like those in the VIA 686B on the XE.

So although I would dearly love to get my hands and CRO onto one, I don't expect it to be this year.

As you say, it's still possible with a lot of effort from all concerned. Best wishes to them all, and if I can help, call me.

_________________
cheers
tony

Hyperion Support Forum: http://forum.hyperion-entertainment.biz/index.php

@GuruMeditation

Quote:

The AGP slot is working in PCI mode.

Excellent news. How would this perform compared to the A1XE's AGP slot?

@jahc
I'm making an assumption here the AGP/PCI slot is running in 32 bit mode, 66MHz. If so, it's 266MB/second bandwidth, which is equivalent to AGP 1x.

The XE in theory I _think_ is a 2x slot, but I'm not sure anyone's confirmed actual throughput on the XEs AGP slot.

So bandwidth may be less than the XE's AGP, or it _may_ work out to be the same is current guess...

_________________
Are we not done with the same silly arguments and flames yet??!

@GuruMeditation

Quote:

As a direct result of the cooperation between Guru Meditation and Troika Partners, we're now happy to inform you about the following changes in the specifications. The form factor has changed slightly, allowing for 1 AGP and 2 PCI slots. The AGP slot is working in PCI mode. Furthermore, there has been radical changes in the bus layouts. Several of the components are now running at 66MHz, compared to 33MHZ for everything before. Notably, the AGP slot and the ethernet chip.

Are the AGP slot and ethernet on the same PCI bus, so thus sharing bandwidth, or not?
I'm assuming this is 66MHz but remains 32 bit for the AGP slot?

_________________
Are we not done with the same silly arguments and flames yet??!

Quote:

jahc wrote: Excellent news. How would this perform compared to the A1XE's AGP slot?

AGP slot in Pegasos II is built around PCI-X bus IIRC. Tests proved that it was faster than ArticiaS AGP "2x". You can expect faster operation than in A1's despite that it's not real AGP. If nothing will be borked.

@wegster

Quote:

The XE in theory I _think_ is a 2x slot, but I'm not sure anyone's confirmed actual throughput on the XEs AGP slot.

around 167MB/sec (peeks maybe at 180MB/sec).

Last edited by Jorge on 19-Oct-2005 at 06:17 AM.

_________________
AmigaOne XE G3/933/VIA/FM801/R200 (fixed),
G3SE/600/Voodoo3/Sil680/RTL8139/SBLive! (noiseless!)
µA1-MK2/G3/933/R200/CMI8738
XE/G4 (broken 7450/800)

@AMiGR

It's always interesting to have discussions where the people involved discuss different matters.

Anyway, they will have to get their beta run out and tested first obviously, and then there is the question about how much testing will be done and how long it takes to fix any problems with the boards.

I'd say there is a good chance that they send test boards no later than next month, and then we will see.

Last edited by Offa on 19-Oct-2005 at 08:03 AM.

_________________
AmigaOS 4 is made of people! It's PEOPLE!

@AMiGR

In what way is any of your rambling a response to anything I said? I specifically asked you to stop jumping to conclusions based on speculations and rather rely on the facts instead.

And the only facts we know of at this point is that they are working on this product, final release date has yet to be confirmed but they are aiming for end of this year.

Oh, and to the statement that you cannot prototype in October and ship end-user products in December: Of course you can. Don't be silly. Are you utterly incapable of working in anything but absolutes or something?

@tonyw
If and when I get some more time to work on my own project you can bet your ass I won't be waiting until my board is finished before shipping dev kits to Hyperion. That would be just stupid.

First you make the software run on a dev-kit, then you make the hardware run the software (trust me on this. I do this all the time at work). Since you know the software works, you only need to debug one of them at the time. I know this is quite the opposite of what we experienced with the A1s, but that's because the SE was the dev kit, and the SE had some major problems

So by the time the hardware starts booting, there had better be a HAL ready, or someone has NOT done their homework.

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@olegil

Right. We shall see about that.

_________________

@all

We have some very talented people who get very excited about working on this project. Can you blame them? The “Prometheus project” is very flexible in its design, and can be scaled to a larger format motherboard. Some of our prototype boards, and features are becoming confused with the Amy’05 s specification.

Quote:

Now, it's important to keep in mind that this is an ever changing process. What you see above might not be the case once the product becomes available for the general consumer. Most of this information you see here has already been posted in the FAQ on the Troika website, but it might be hard to get a good overview of the changes to date.

@AMiGR
Quote:

BTW, there has been another change in the specs anyway... They are adding an AGP slot and changing more than the size of the PCB. Now, tell me that a December release is still even remotely realistic...

Very thoughtful post but wrong. The Amy’05 specs located here

Quote:

They have changed the size of the board to 190x170 in the comparison chart.To me this indicates that they just moved the PCI slots so they both can be used in a normal mATX case.

Yes, and in future.FAQ we will explain.

_________________
Troika NG progress reports & news releases will be released as such, from my office.
ttp://www.troikang.com/ Amy, A NEW HOPE
============================================
Opinions I might express are mine alone,and do not represent those of Troika

I suppose flexible design could mean that those who whishes a higher specced board could have one sooner than they think if you have success with Amy'05.

_________________
AmigaOS or MorphOS on x86 would sell orders of magnitude more than the current,
hardware-intensive solutions. And they'd go faster.-- D.Haynie

@AMiGR

See about what? What on earth are you on about? I made absolutely zero predictions in any of my posts, what is there to see?

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.

@Troika, who said:
"Some of our prototype boards, and features are becoming confused with the Amy 05 s specification."

This is getting VERY confusing. So you are saying that what Guru Meditations posted previously:
"The form factor has changed slightly, allowing for 1 AGP and 2 PCI slots."
Has nothing to do with Amy '05? You could have fooled me!

If someone is going to post info about Troika prototypes OTHER than the Amy '05, then perhaps they could make that much clearer next time. Because I think that as far as most of the public is concerned, there is only one Troika motherboard!

_________________
Author of the PortablE programming language.
It is pitch black. You are likely to be eaten by a grue...