Click Here
home features news forums classifieds faqs links search
6071 members 
Amiga Q&A /  Free for All /  Emulation /  Gaming / (Latest Posts)



Lost Password?

Don't have an account yet?
Register now!

Your support is needed and is appreciated as is primarily dependent upon the support of its users.

Main sections
OS4 Zone
IRC Network
AmigaWorld Radio
Top Members
Amiga Dealers
About Us
Terms of Service

IRC Channel
Ports: 1024,5555, 6665-6669
SSL port: 6697
Channel: #Amigaworld
Channel Policy and Guidelines

Who's Online
35 crawler(s) on-line.
 55 guest(s) on-line.
 0 member(s) on-line.

You are an anonymous user.
Register Now!
 matthey:  7 mins ago
 amigakit:  10 mins ago
 OneTimer1:  14 mins ago
 BigD:  15 mins ago
 NutsAboutAmiga:  1 hr 44 mins ago
 zipper:  2 hrs 32 mins ago
 Frank:  2 hrs 46 mins ago
 Lou:  3 hrs 40 mins ago
 bhabbott:  3 hrs 42 mins ago
 MichaelMerkel:  4 hrs 17 mins ago


Main »» Interviews

Interview with Alan Redhouse of Eyetech (21-Nov-2003)  Popular
(Read 51594 times)
AW Staff: Hello Alan, thanks for sponsoring AmigaWorld and for taking the time to answer these questions!

1) Here at AmigaWorld we are now providing Eyetech's Official AmigaOne Forums, News and Updates. Can you share some of your future plans with regard to AmigaWorld? For instance, will Eyetech partcipate regularly within AmigaWorld's forums?

Our main objective - and therefore the main focus of my time - is to promote the AmigaOne hardware and OS4 into broader markets. This means that, despite my best intentions and wishes, the time available for my direct participation in the forums (or is it fora? it's a long time since I did latin at school) will necessarily be limited. I'll continue to read the debates and put in selective comments where I think it is helpful, relevant and important to do so.

Not having more time to actively participate in the forums is something I really regret, but the bigger picture is the most important thing to concentrate on right now. In the interests of the existing Amiga community we absolutely have to broaden the market for the A1 technology and for OS4, even if that involves using Linux as a trojan horse. And when I say 'we' - meaning the broad alliance of Amiga Inc., Hyperion and Eyetech - in practice the other commitments of Amiga Inc and Hyperion means that that responsibility is, for the moment at least, mainly falling on Eyetech's shoulders. This is fine by me as Eyetech's historical expertise is selling big-picture technology-based solutions at corporate and governmental level. This need to focus on these bigger markets is the main reason that we've handed over reins of the UK retail market to Sven Harvey's company, Stellar Dreams.

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?

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.

3) Many people state that they will wait with buying the new hardware until AmigaOS4 becomes available. Do you expect many more sales when AmigaOS4 is released?

Many, many individuals have contacted us and the individual A1 dealers to say they will buy an A1 once OS4 is deliverable in end user format. However, since OS4 running on the AmigaOne hardware has been widely demo'd since the beginning of October we have seen a new surge of AmigaOne orders. My guess is that this is because these purchasers can see that the end-user version of OS4 is close and they can still, just, take advantage of the Earlybird offer, thus saving themselves the cost of OS4 when released. We are also working with Hyperion to try to make a hard-disk installable version of OS4 beta available to all existing A1 Earlybird customers and developers, hopefully before this Christmas.

Many other Amiga supporters though are naturally suspicious of promises (not surprisingly given the number of false starts over the last 10 years) and will only buy when they have personally touched and felt OS4 on the AmigaOne hardware themselves. At that point dealers are indicating that they are likely to be overrun with orders.

Inevitably there are a few who email us to say that they will only get an AmigaOne when hell freezes over. That's OK by me - everybody has the right to choose. But we really do have to focus on what is important now, so this last group may have some time to wait as high performance refrigeration projects are not a currently high priority item for us.

4) Next to the current Amiga market, at which markets are you planning to target the AmigaOne hardware?

See 2 above

5) A pre-production version of the new MicroA1 board (Mini-ITX/170mm/170mm) has recently been demonstrated running Linux. What changes can we (i.e. onboard graphics chip) expect for the final production model?

The MicroA1 is targeted at the price-sensitive market where ultimate hardware go-faster badges are of lesser importance than the ability to provide a cost effective solution. We will therefore use the graphics chipset which give the lowest cost of production consistent with the best OS4 support. This will almost certainly be the cheapest Radeon chipset that is well supported by OS4 and LinuxPPC at the time the MicroA1 goes into production. 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. This is to keep the cost of the basic board down for industrial applications which do not need these capabilities. This standardisation will allow us to benefit from economies of scale on the basic board, without inhibiting expansion flexibility.

6) Are there any plans for an AmigaTwo (not necessarily in the near future) that would be produced if sales of the AmigaOne were successfull, with the features of the MircoA1 (USB2.0 Firewire etc) and maybe able to take IBM PPC 970 but with 4 (or more) PCI slots?

The 970 chip will not be available for general sampling until well into next year and realistically not generally available at consumer-compatible prices until the end of 2004. Because of its different architecture it will require a completely new board design. This will almost certainly be initially targeted at the professional/sever market and so will, of course, come with all the I/O bells and whistles available at that time.

7) In light of the difficulty in now purchasing grapics cards for AGP 4X slots or less, are we likely to see a modified board incorporating an AGP 4X/8X slot? Indeed, would there be a possibility of a future PPC Motherboard incorporating the PCI-X slots for even faster graphical transfer?

The Articia S northbridge used in the AmigaOne supports 2x AGP speeds. In comparison, the Marvell chipset only supports PCI-X mode - assuming the board using it uses PCI-X slots of course (which are different mechanically to normal PCI slots). 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. True AGP - as used on the AmigaOne - achieves its speed by, for example, sideband addressing - allowing the graphics card to request data without interrupting data flow on the main bus - and pipelining. This is of course an over -simplification - it would probably make a useful AmigaWorld forum discussion topic, allowing people far more knowledgeable then me to comment on the nuances of the AGP architecture.

In hardware terms however, virtually all AGP cards which advertise speeds of higher than 2x, are 2x compatible - provided that they are designed to operate electrically on the voltages present on the AGP 2x connector. Having said this the AGP speed factor on a true AGP bus is very much over-hyped, in my opinion, as far as delivered video performance is concerned. For example, the actual delivered speed difference between, say, 2x and 4x AGP is generally accepted to be only around 15%.

This is because all modern graphics cards have huge processing capabilities in their own right. In all high framerate 3D games - where graphics performance is pushed to the limit - the AGP bus is therefore used, in effect, to send an executable program to the graphics card which then generates the display data. This is a relatively low bandwidth data transfer function on a true AGP bus, and one which therefore does not benefit significantly from higher AGP speeds. It is the faster, more capable graphics processor that mainly improves performance in 3D games, not the nominally higher speed of the AGP bus. In effect the 'AGP speed' quoted on the outside of the box is really just a simple label for impressionable PC users and not any objective guide to graphics performance of the card.

In addition the AmigaOne has a major inbuilt bandwidth benefit compared to normal PC architecture. This is because the transparent synchronous design of the Articia S allows direct memory-AGP transfers without suspending cpu-memory operations as is required in the x86 world. Just ask any user who has installed the same version of, say, PPC Linux on an 800 MHz AmigaOne and a PC of 3x that clock speed as to which is the most responsive. The A1 wins every time.

8) We have seen a number of people asking about timeframes for delivery of their systems/motherboards. If someone were to place an order tomorrow for either a complete G4 system, or a motherboard at an official AmigaOne dealer, what kind of delivery time would they reasonably be expecting to see?

A1XE motherboards can now be delivered ex-stock from most serious A1 dealers. However this Autumn's local shows have had a habit of exceeding demand expectations and have cleared most dealers out of stock. So it is probably reasonable to expect a 2 week lead time on bare boards and up to twice that for prebuilt systems.

Questions from AmigaWorld vistors:

9) royleith: There still seem to be questions about reliable operation of DMA with the Via Northbridge. Is this finally sorted with UBoot, yet, or is there more work to do?

There have been two problems with DMA, neither of which is down to any bug in the Articia 'S' northbridge. The first is associated with initialisation of the VIA southbridge used on the board and the second with Linux drivers.

The biggest problem was that the VIA interrupt controller was missing interrupts on high loads, such as those generated by DMA transfers. The solution turned out to be a requirement to program an undocumented register in the VIA southbridge. VIA were particularly unhelpful in coming forward with documentation about what was obviously a well known and solved problem in the x86 world. AmigaOne developer Adam Kowalczyk working together with Bill Mueller 's team at MAI Labs sorted this out by some significant inspiration and a great deal of perspiration.

The second problem is that many PPC-ported Linux drivers (which we must not forget were developed for the x86 environment) have often implicitly relied on the fact that Intel architecture prevents simultaneous memory access by the CPU and a PCI bus master. In the AmigaOne both processes operate simultaneously and transparently to each other and so some of the ported drivers fail under extreme pressure. IDE DMA drivers have been modified to make them behave properly on the AmigaOne and a universally applicable OS patch is now in test. 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.

These issues are exclusive to the PPC port of Linux and are irrelevant under OS4.

10) Eric_Z: With the release of the A1-lite (MicroA1) approaching, are there plans to update the A1 (new northbridge/more features) so that sales cannibalisation is kept to a minimum by further differentiating of the two products?

The two boards will be differentiated by their physical size, expansion capabilities and cost. Of course we are also looking at several possible design specifications for future products on an ongoing basis, but our main priority is to expand markets and get volumes up. A $100,000 board redesign process costs $1 per board spread across 100,000 boards, but $50 per board across a 2000 run.

11) Eric_Z: When will we see a DDR-RAM cabable and AGP 3.0 compliant northbridge in use amongst your product line?

When it becomes cost effective to do so, taking into account chip capabilities, availability and pricing, and the market advantage, if any, that that such a redesign would give.

12) Chunder: Do you have a "roadmap" illustrating the direction in which you'd like to see the A1 development going?


13) Bobsonsirjonny: With the MicroA1 are there any plans to house it in its own unique case? I, and many others feel this is very important for recognition and branding that it has its own face. Its own identity which people can become attached too.

The real question is not if it is feasible - it certainly is - but who would buy it and at what premium. For example if a custom case cost $200 more than a standard case (which a nice design easily could do in low run quantities) how many people would choose to spend their money on that rather than enhanced hardware options? Of course when we get back into the mass market a little way down the track custom cases will be an absolute requirement, and probably also the most economical option anyway.

For now however we are working with industrial design house and Amiga enthusiasts Naya Design, to introduce a range of AmigaOne case 'mods' which are relatively modestly priced in themselves, but which will turn a decent off-the-shelf PC ATX or ITX case into a distinctively Amiga product. Watch this space!

14) Chunder: The A1 (and microA1) have been touted as being solutions for the embedded market - especially the microA1, I'd expect. Would you anticipate such solutions running AmigaOS, or Linux?

Either, which ever is most cost effective for the application. See 2 & 5 above.

15) Vidar: How is Eyetech's policy for replacement of write protected BIOS chips which cannot be updated without proper equipment? Would it be possible with some kind of swap arrangement in order to minimize downtime for direct customers? i.e. Eyetech provide new chip, customer then return the protected one?

We have tried the 'send out the new chip and wait for the old chip to be sent back' approach. In most cases we are still waiting - I guess its just human nature to make the return low priority when the original problem has gone away.

Most dealers now have ROM reflashing/unprotecting facilities, and will make a small charge for this service. But remember that the original A1 specs said that the ROM would be non reflashable (in fact an PROM). Making them reflashable was a later change after different anti-piracy mechanisms were built into the A1 boards.

16) Swoop: With the Aone lite being announced for 1st Qtr 2004, what are the chances of an Aone Laptop??

Give me an order (and a 50% advanced payment) for 100000 units and you can have one (or any number up to 100000) by next March. More realistically this is only a product that will be considered when the A1+ OS4 is well established in the broader marketplace.

17) Kay: In the last Total Amiga there was some information about the dual 1.3GHz CPU module being planned, and I must admit that it sounded a bit too expensive for me. How about making a cheaper module with a single 1.3GHz processor?

The single module is trivial to do - we already have 1.1GHz 7447 modules and in fact one was seen on the MicroA1 at the OASE show in Graz this month. But we would still need around 400 advanced orders to make it worthwhile at this point in time.

18) Bodie_CI5: What results do you think could be yielded from the trip to China you recently embarked upon? Do you think that you will be able to make any further inroads in the greater South/South East Asia Region?

The conference delegates were expecting to hear exclusively about Linux, but I hopefully presented OS4 as a very cost effective alternative in specific application areas. There was a lot of interest in seeing it working as a 'proper' GUI-based OS with a mature selection of applications, and not just as another limited functionality embedded OS. I can honestly say that the overall potential is enormous, but what the exact target markets are and how we intend to sell them is not for a public forum, sorry.

On my next visit I intend to make further inroads into SE Asia by the usual methods of public transport, hire cars and possibly the odd rickshaw.

19) Toaks: I would really like to know at which price you have this product aimed at (amigaone lite/amigaone micro)?

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

20) Eric_Z: I am wondering, what is the status of the Micro ATX A1 (not MicroA1)?

The Micro ATX board is scheduled to come out later next year, after sales of the MicroA1 and A1XE achieve sufficient volume.

Hope this helps.

Thank you all for your ongoing support and patience.

Alan Redhouse

View/Post Comment(s)
- MikeB
[ home ][ about us ][ privacy ] [ forums ][ classifieds ] [ links ][ news archive ] [ link to us ][ user account ]
Copyright (C) 2000 - 2019 was originally founded by David Doyle