Quote:

Well, fact is, tk@bplan ensured me I will get back a working peg2 board within two weeks. After 3 weeks I gave them a phone call asking about the status. It has been told to me, that the firmware update will arrive "tomorrow" and my board will get send back to me the day after that. A week later I was calling again, still asking where my board is and what's happening. I needed that board to reinstall some SGIs for Debian Developers (namely Matthias Klose doing some gcc porting work for mips), because the peg2 is my SOHO server.
Anyway, I was told that there were some more fixes, but the firmware update will arrive by the end of that week or maybe the beginning of that coming week. So, I told them I need my board back for doing work (s.a.) and they should sent it back to me by the beginning of that next week, nevertheless if the update arrives in time or not. Firmware updating is not that big problem, when nothing else needs to be done (like exchanging hardware).
So I got my peg2 board back. That was two weeks ago and yet still no firmware update around, although it has been told to me that it will be there "in the next few days".
The latest information I got was from Vesalia telling me, that the firmware update will be made available on 28.7., but the guys at bplan are currently not reachable. . o O ( uh? maybe holiday time... )

We'll see if bplan/Vesalia/whoever is holding promises or fooling around, if the update is available or not on the 28.7.

regards,
ij

Looking at the current crop of business partners, Genesi needs to get its hardware and software used in solutions, where the margins are high or find a volume resale for software that also can be sold at a high profit marging to a high volume of people.

What do we have?

LinuxPPC sold to end users
-- most people will question why they want LinuxPPC, I dont see a good reason to go LinuxPPC rather than Linux x86 at the moment
-- low volume potential

MorphOS sold to end users
-- Why MorphOS and not Windows or Linux? Those that it makes sense to are probably already in posession of an AmigaONE or a Pegasos.
-- very low volume potential

Peg/LinuxPPC sold to Freescale
-- for evaluation purposes and chip testing?
-- very low volume potential

HP allowing Genesi to bundle their software with Peg/LinuxPPC
-- is this that much of a big deal?
-- who will buy it?
-- very low volume potential

IBM allowing Genesi to use its Ready For IBM Technology on their boards, after self testing with their G3s
-- so?

IBM allowing Genesi to be listed on its solutions database list
-- who is looking?

So Ive excluded the ones that have gone by the wayside.

Actually the last one is the most important one, because the /only/ way to make money is to get your system used by at least one solution provider, and for it to work, for you to get a service contract and for your systems to be used in volume in that solution. Then, in theory, if your kit does not suck, it will be recommended by the ISV and potentially reused - you have a testimonial.

This is far better than toadying up to your suppliers, they can't introduce you to ISVs in a credible fashion.

You need to toady up to ISVs, in order to do that you need to get introduced to a few through the "network" and in order for that to happen you need a good sales story, and in order for that you have to have demonstrated something in their business sector on your systems, so they know you can provide a platform that they can use at the right level.

This means middleware.

List the middleware that is available on linuxppc, and who uses it and what for, and what its compatible with, then you will see the available markets for you. Then hire someone to spend 6 months coming up with a demonstration package and material for the most lucrative market you identify.

Then start flying around all over and showing it off. THEN people will take you seriously, THEN you will get the margins needed to sustain Genesi while MorphOS is developed.

If I was a red troll, I wouldn't tell you this business 'secret' I'd carry on watching you flounder.

Move your marketing strategy to the right, away from hardware, and make the hardware part of the solution. Then you might actually be part of a big installation.

Quote:

True, but when I ask them, what's the status or when will it be ready, then I expect answers on which I can rely. This hasn't been the case so far.

Quote:

Yeah, I have heard of such uptimes from svenl as well and that's one of the reasons why I believe that my board is what we call a "Montags-Board" in German.
Nevertheless I wanted to buy a (stable) working board which I currently doesn't have, despite of how much memory I use (512M or just 256M) or from which manufacturer.
There's no difference as well which kernel I use... as well 2.4.x panics as 2.6.1 to 2.6.7 did and does, again it doesn't matter if those are precompiled by Sven or self-compiled by me.

We'll see if that upcoming update will turn my peg2 into a stable machine... ;)

regards,
ij

Quote:

As broadband connections gets more and more a commodity in the every mans home, there is a growing demand among some populations to set up servers of their own. This can be for file sharing services like Direct Connect, FTP, Webservers, gameservers, etc. Windows is an extremely poor server OS; no matter if you download all patches the same moment they becomes available, it's still full of holes and after a couple of months you will have to reinstall the whole thing. Linux is a good server OS (if you have the patience to learn it). But then there is the noise problem. A server often needs to be connected 24/7 to serve its purpose, and many people are living in apartments, some in large apartments where they can put their computer in a closed room, but others lives in small apartments where they perhaps have their computer in their living room or their bedroom. Noise can be a problem there. Take a Pegasos2 G3, a Seagate Barracuda, a silent Nexus PSU and you have a server that is *totally silent*. You can even put it under your bed, and it won't disturb your sleep. You will have to check the Power LED to see if it's actually powered on. The only x86 counterpart I can think of is the VIA Epia motherboard, not with the C3 processor, but with the fanless Eden processor. While that one is cheaper (the low price being its only real advantage), it also is quite weak when it comes to performance (especially if one would compare it to a fanless 7447A G4). The MiniITX form factor will also limit you if you want to add PCI hardware to your server, especially if you are using the computer as a combined router and server and thus needs to occupy the only PCI slot with an extra Ethernet card. No way of adding extra IDE controllers and such. The Pegasos2 has everything it needs to be used as a router, out of the box, while still leaving three PCI slots free for future use. Furthermore, I can't think of a single mainstream x86 motherboard that has Gigabit Ethernet implemented the way the Pegasos has; directly into the Northbridge and not through the PCI bottleneck. A fast Gigabit network is extremely nice to have in your LAN, trust me!

I have showed my Pegasos server to some friends of mine, who are quite seriously into what we can call "file sharing" (hrmm ), and they all were impressed. They themselves have had to rebuild wardobes inside walk-in closets to function as little server rooms (and similar), in order to avoid being thrown out by their girlfriends because of the noise their x86 servers makes. No need to say they were quite impressed by the Pegasos, both by the performance, but most by the silence. The price is not an issue to them; the current price tag is not *that* high really; they all have jobs, they take their computing hobby seriously, and we all have x86 machines that costs a lot more than the Pegasos. What I think it's stopping them is more a feeling of uncertainty; the Pegasos is young and unknown, it's hard to find if you want to buy one, and not a lot of info is available on a broad scale (except from pegasosppc.com and such). They think the Pegasos hardware is *excellent* as a server box, they *love* the silence, they think the price is acceptable, but they still hecitate, probably because of the "un-established" status of the Pegasos. (BTW, the "IBM" logos and such things you mentioned may be small components in a process of lowering this uncertainty; familiar logos alone are of course not enough, but it may help together with other things).

Just out of PURE curiosity..

.. how much would you pay for MorphOS? I mean, for your first update, and then subsequent updates? I tentatively start the bidding at $30 for a full release scaling to $20 per major update (from previous version only) with free incrementals?

.. in the above scheme, 1.5 would cost $30, 1.6 would cost $20 because you already have 1.5, and 1.5.1 would be freely available as an update. If you skipped 1.4 to 1.7, you'd have to pay $30 for 1.7 (which is cheaper than getting each upgrade as you go along, but would you really want to miss 2 versions? :)

MorphOS definitely needs a "service pack" scheme (some interim developments and fixes.. downloaded and installed in a batch... like a BoingBag but with a name that doesn't sound like a low-grade porn movie prop). Downloading individual files from Zargos is going to go away if I have anything to do with it (and note.. I *DO*)

This isn't fixed or even considered, it's only out of my own crazy brain, I'm guaging opinion..

Neko

Hi Neko,
Lot of users are ready to pay Mos to get faster release.
IMO isn't the good solution. We all want Mos to be use by new user ( i mean no ex amiga user) Mos isn't at the stage of Windos or Mac for now. I believe Genesi startegy is to make Morphos known by a maximum of user (ie via linux). Leaving it as a free OS will bring much more interest.
Some of us are frustrated because of the poor information release by the mos developers and/or the time it take to have new update, but more people will use MorphOS more update/soft will come.
Let's be patient and see what happen the next couple of month. Genesi announced
"A MAJOR IT COMPANY HAS AGREED TO RESELL MorphOS"This will boost all development ! ;)

Well, I know that it's a little strange and MS was flamed to death for it
But I would like the idea of a rent-a-os thing ... sth like 5-10 Euro
per month for one or two years until MOS is in a competitive state
compared to MacOS,Linux and XP ...

For a single major update every six month I'd readily pay 30 Euro.

Well in this case, they should add a "donation" section on Morphos web site and people who want's to give money to help the Mos team can do it :)

@Neko

The prices you mentioned would be acceptable for me, as long as the money goes 100% to the MorphOS crew - so if you'd really decide to go that route one day, I'll happily pay.

The suggestion of a donation possibility similar to the AROS bounties would indeed also be an alternative I'd support, but paying for updates seems to be the better solution IMHO.

[ Edited by Senex on 2004/7/23 13:39 ]

Another busy day and another quick note...we will try to catch up on all this over the weekend.

1. Why PPC instead of x86:

Selling the Competitive Advantages

2. Here is some text from the Pilot Project master document:

The thin client/file server configuration is the anchor of the Pilot Programs. It is from this point and beginning with the user that the PowerPC will be extended into the entire enterprise eliminating the need for multiple platforms and multiple operating systems. The ultimate goal is to collapse the entire IT infrastructure into a more easily manageable solution and demonstrate that one platform and one operating system with a complete open source software stack can fulfill the needs of users and companies.

We seek to address the needs of the growing thin client and file server application markets. Our intent in the Pilots is to develop the supporting documentation we need to effectively make an argument for our offering. Nevertheless, the success of our offering should not be focused on the PowerPC or even the PegasosPPC, but the total solution we are providing. Our proof points are focused toward a final objective of providing a total system solution and not simply the means to accomplish it. CIO's only seek to maintain the IT environment that they need and lower their costs. There are very few (if any) strategic applications or environments in the market we are attacking. We are not going after the niche high-end specialized applications used in certain office environments - rather we are planning to address the market segments with well understood application set - like those of the thin client space. We are not going after the departments that use Apple computers. We are going after the other 85-90% of a company's users.

Today, many of these factors referenced above are usually not included in TCO calculations. An objective of the Pilot Program is to identify metrics that will establish new criteria for performance and new standards of performance measurement. Long the hallmark of the PowerPC, lower power requirements, lower operating temperatures and thus better reliability, have been ignored by the broader IT market. Our task is to innumerate, document and market these virtues in the context of normal day-to-day computer use.

The goal of the Pilot Projects is to support the introduction of the PowerPC based Open Linux Platform we are launching. The Pilot Programs will validate the following:

The Linux/OSS/PowerPC environment lowers TCO vs. X86 Lintel/Wintel alternatives across a spectrum of configurations from thin client to cluster.

DaveP has the right ideas and we have been working towards them quietly for months. You will be reading about it in a newsletter from one of the companies you mentioned soon...

More later!

R&B

P.S. AGAIN: We are 100% in favor of continuing to develop MorphOS. For the moment, things are slow because it is just a matter of resources and time.

Two more things...

1. Gerald and Thomas made boards yesterday. That is why they were not at bplan.

2. The next OF upgrade is being tuned now. Gerald, Nico and Sven have done great work.

R&B

Hi Neko,

This is a very good idea in my humble opinion.
These amounts are also acceptible.

A simpler option could be:
Free latest MorphOS version with new Pegasos and after that paying a fixed amount for each update or so.
In this way all new users get a free copy of MorphOS to try out.

Bye,

Spidey

Quote:

Honestly, why are you telling this to us?

We are PPC users alredy We have Pegasos already, we develop for MorphOS already, we use MorphOS already. Aren't it right time to knock doors at linux/x86 forums and ask what they want? They are the target after all

We are!

We are just sharing the process and discussion with you because we want you to understand what we are doing and why.

R&B

OK, so, after I thought about it for a while, and read the other comments, here are my 2 eurocents...

First of all, I agree with those who pointed out the fact that 1000$ for a development system with support ought to be next to nothing to most businnesses. OTOH, Neko spoke about getting some recognition from thousands of developers on a weekly basis (in the most optimistic case). So I think he was speaking about "small" developers that might get interested in MOS. BBRV, instead, were speaking about major partners, about whom we don't know anything yet. I'd bet that any major partner has the possibility to buy a Pegasos, and I think that Genesi would even offer them free boards, if there is some concrete interest...

Anyway, if there is a "strategic" need to have a MorphOS demo available on PearPC, so be it. PearPC is rapidly evolving, and while it might not be half as good as running MOS on native HW, it has already reached an usable state. I saw that one of the most reported critics was that MOS isn't really ready for prime time. I think this is an overstatement: MorphOS definitely lacks a good web browser, and in my opinion, also a good packaging (i.e. while a native TCP/IP is not integrated, some form of emulated one - even a demoversion would do - should be there, as well as some of the freeware "essentials"), and some features are still not implemented, but it already is a good OS. I know several people using it as their first (and in some cases, only) system. Of course, if you were going to show it to the "others", you would need most of the 1.5 features (like the updated Ambient, since the default handling of different filetypes is a "must have" for everyone out there). But then I'm pretty sure you would have one of the leading "alternative" OS. And then, what would I show? Well, I would show what it can do, and basically it can do most things an OS needs to do: Internet programs (aside the well known browser problem, MOS has some small gems, like AmIRC, YAM, AmiTradeCenter, ...), Multimedia support (how many filetypes are not recognized by mPlayer or Frogger? and with the aid of plugins like Euphoria, doesn't AmigaAMP or ANR look very good - well with some skins ANR looks very good by himself?), the high degree of configurability, USB support (hot plug of an USB joypad while using your favourite emulator), and all the rest (applications, games, demos). If things are not running at full speed, it's not a problem, I think the point is to show that MOS is already at a good point. And the Amiga compatibility as well, because MOS has an advantage against most new OSes, and it's that the MOS user can already access to hundreds of SW titles, something none of the other alternative projects had since the beginning (and it's what hampered them, in the first place). Of course MOS is not at a point in which it can be compared with Windows or MacOSX or Linux, in terms of third party support. But then, whom else can?

Honestly, though, I don't have great expectations from all of this. But there is definitely the need to get the word out. As I had the possibility to attend at a MOS showing to people outside "our" community, I can tell that the general consensus about it is that it is already very good. Sometimes people can't believe at the speed some operations can be performed with a simple and "slow" 600MHz (like handling PDFs). After all we would not be here if we wouldn't believe MOS is a good system... far from perfect, agreed, but very nice indeed. Although we're all waiting for the 1.5 to be released before hell freezes over!

Kind regards,
Andrea

Hi poundsmack,

It's true we hear a lot of anouncements of exiting stuff and it's also true later on we don't hear anything of somekind of progress or so.

Let's hope, like you ended your post, that Genesi will choose a clear direction and will follow it.
I do get the feeling Genesi's delivered a very good product and hadn't thought beforehand what to do with it (vision, mission, target, etc. The normal ISO9001-stuff ).

Don't give up faith. let them choose a path first

Edit:
I just saw bbrv has replied and in the post is a clear direction mentioned.
I only hope if this goal is not too high to reach


Bye,

Spidey



[ Edited by Spidey on 2004/7/23 15:10 ]

@guruman;

Yes, although I think people are mindfully confusing the issue of discovery in a software development project, and predetermined development.

What I am talking about is a SIGNIFICANT cost saving (let's say $30 instead of $3000 :) on test driving MorphOS to see if it is suitable for development.

I am not talking about saving money for companies that have already pledged or are to be coerced immediately into developing for MorphOS. They have already decided and are going to be in the phase where they can lay out HUNDREDS of thousands of dollars for development tools.

For the initial stages of a project, where you decide.. hmm.. do we use ucLinux or VxWorks or MorphOS? Buying an ODW box makes little sense just to determine that. VxWorks and ucLinux already run on x86 so booting them on your PC or in VMWare is a simple task - and you can determine then whether you want to use them. MorphOS requires dedicated hardware to test and runs the risk of being instantly discounted because of it.

For embedded developers choosing processors, the major aspects they are faced with are power consumption, heat characteristics, overall functionality and of course availability. It's generally not "how fast does it resize a window", although that functionality alone is what impressed the hell out of some academics wanting to produce clusters - the responsiveness of MorphOS on our hardware proved in concept the efficiency of the hardware itself.

There are lots of PowerPC solutions on the market from full blown desktop-style processors (G4) to SoC (IBM 440GP) to highly embedded signal processing and communications (PowerQUICC). MorphOS can run on all of them, if so needed, and in general the processor is already chosen; all that remains is what OS to run on it.

At the lowest level, we would tap into that market by having it offered, resold as an evaluation package, alongside the full-fledged ODW and individual Pegasos boards.

MorphOS under PearPC is not going to be slow. I have been running MacOS X on my Pentium 4 here and it's definitely more than usable - given the incredible performance requirements of MacOS X, that is really quite impressive. We all know MorphOS is much more efficient than that! With work, PearPC could provide all the acceleration features MorphOS could want (CyberGraphX interfacing with the host OS). Remember we're pushing our own OPA which is PERFECT for this, in context.

Neko

Just to contradict Framiga I'm happy to tell you all that UPS just half an hour ago delivered to me a box from bplan with inside two defective boards replaced by new ones!
We waited a bit more that the 2 classic weeks you spoke about but at he end we received two completely new boards.
I'm sure my waiting customers will be really happy to hear about this...

Quote:

Well, your goal is to reach companies. Then consider again to give Pegasi for rent - 4weeks/100 EUR. As said this is not to raise funds for gensesi, but may compensate costs for genesi. 100 bucks are really *nothing*.

And it gives you so much more credibility when you can offer them just to ship a box for testing for a little charge. I test a lot of eqipment at university, expensive equipment (mostly optical stuff), most of those companies offer a test rent for a little charge.

As long as we speak of companies the demand will not be 10000 (add or reduce some 0s here) maschines, but only some. Just start with that in a small scale, assemble say 20 - 50 maschines and start the offer, if noone will take the offer then you still can sell the maschines... It's close to riskless. According to BBRV boards were just produced yesterday, the offer might start within 10 days!

Quote:
Or on Qemu...
http://fabrice.bellard.free.fr/qemu/benchmarks.html
You would have the speed (and hopefully responsiveness) of about a peg G3 600 then in emulation (rough estimate a factor 4 to 5 lost by its emulation, run on a 2.2 GHz AMD64).


[ Edited by azalin on 2004/7/23 16:23 ]

I wonder why this "major company" is such a secret. Sounds a lot like the usual PR-stuff to me (Atari-name, Pegasos-gameconsole etc.).

And PearPX IS currently very slow. I dont see any sense to show MOS slowly emulated on a x86-machine. Even with a 3 Ghz-PC you would be lightyears away from a Pegasos II with 1 Ghz (an PearPC only emulates a G3, too). Which customer will be attracted by a slow niche-OS in beta-stade? Better show it as fast as it is on a PegaosII, with improvements (MOS 1.5) and more MOS-software. That could be attractive.

Quote:

Well I and I think many others would have been a lot more happy if you would have just said from the start; "Hey we have no money at present and will be selling linux systems to help get some. At this time MorphOS is on hold due to a lack of interest and cash flow from it, soon we hope it will return to full dev" Thats all we want, and if we have to send that cash to the dev team I think the other various threads have shown we are all fully ready to do just that. Most of us dont want linux, we dont want windows. I got a Pegasos for MorphOS and SOLD my long loved A3000 to do so. I have thousends of dollars in that machine and took a gamble. If it was for nothing I will be crushed.

MorphOS dev tem:

Get an address and a point of contact together so we can start sending you money. If Genesi cant/wont do it then we will take it into our own hands and support your coding time.

Well on thing is for sure, BBRV keeps surprising us at every moment!

Many things were said, and many think emulation is the evil in itself... gee! You have a computer and use it as a commodity, people doing it everyday with windows, and we all know how slow and painful it can be.

One shouldn't forget software drove hardware and hardware drove software, it's the old chicken 'n egg paradigm all over again. By providing more compelling solutions where you can run MorphOS, you're breaking it up.. it get a wider exposure in other ways otherwise it could never had, you could have one company that by 'magic' could acquire at almost 0(zero) expenses as many developer machines being those the ones they already have, and using Pegasos for testing proposes only for example...

Things can be made well... critical drivers could be made to be used like Amithlon and used natively, you would never get a 'bad impression', because: first it wouldn't be the real thing so while having the ability to test the solution presented to him he would know it could never get the performance of native. If one user sees a Photoshop beater on MorphOS running and while trough emulation it seems to run on par with t in user responsiveness, it could well buy a system where this app would run native.. on the other hand, without having seen it he would never buy it.

Amithlon has shown up that an emulated solution can be made faster then some of their compelling solutions natively (compare it with windows on the same machine), had it had one 'exclusive browser' for it and ironically it would be ahead of all other Amiga native solutions.

As for screenshots and such instead of... it's never the same thing as using and touching, there are things which an Amiga like OS does fast not only because of how the OS itself behaves but also due to how GUI manages to do it (screens, assigns, etc...).

In fact you could well one 'slow' solution running way faster then native! I do many things faster on UAE that I couldn't do on Linux natively due to this.

Are you implying that users will ditch their pc on the trashcan(tm) just for using Pegasos?? I find hard to happen until this software gap is filled... so I'm for all with BBRV on this thing!

pixie- writing from a paradise called portugal

PS- you have to run UAE in Linux, seeing is believing!

@bbrv

Okay.

Anyway, PearPC itself is not the worst possible idea for evaluation purposes. If it was hardware it must be sent physically when an emulation box could be transfered electronically in seconds.

There could be even ready to use MorphOS demo available for download for x86 world (Windows). Another matter is how well PearPC works, I never looked into it.

PegRent could be suitable too for those wanting to see the HW.

@itix

Quote:

While I never saw it running myself, sayings are that he performs everyday better!

[ Edited by pixie on 2004/7/23 16:48 ]