Hi,

I'm considering funding the design & production of a new PowerPC system (well, the motherboard, the rest are typical pc stuff and a case). No this is not a joke, I've been wanting to do this for a long time, and perhaps the chance will be given to me now. But before I spend any money on this, I want to do a little market research first. I know the market is literally "dying" for a new powerpc motherboard, but exactly how many are there that want to buy one?

Ok, let's give some rough specs first. I'm considering 3 choices -not in order of probability/importance:

1. MPC8640D-based. It will be dual core at 1Ghz -most likely, higher frequencies are much more expensive and the cost of the final board would be prohibitive.
2. MPC8610-based. Single core at 1Ghz, slightly less expensive, and includes a 2D DIU display unit -quite fast, but no 3D unfortunately.
3. QorIQ P1022-based. Again dual core at 1Ghz (1055Mhz to be precise). Apart from the much lower chip price, this one includes dual gigabit ethernet, dual SATA, USB 2.0 and a 2D DIU display unit (same as the MPC8610). So this one would lower the cost of the board quite much. Disadvantages: No AltiVec unit (it sucks I know), though it includes an SPE unit which is not that bad, and availability will be in Q3/Q4 2010, so that's a long wait.

Now, the end motherboard will probably be MicroATX (in the 8640D/8610 case) or PicoITX (in the P1022 case), and it will definitely include:

* SATA connectors
* USB (possibly 2 back and 2 front, but that's discussable)
* Dual gigabit (at least one will be there, in the case of the MPC8640D we might even have 4!!!)
* Sound (of course, SPDIF support will definitely be there)
* 1 PCI-e slot 1x
* 1 PCI-e slot (4x in the P1022 case, 8x in the MPC86xx cases)

Ok, what I want to know is if people would really really buy one of these. End price is estimated to be ~around~ 350EUR for the P1022 board or ~500EUR (definitely more in the case of 8640D) in the case of the other boards. Besides being more expensive, the MPC86xx chips, don't include SATA, USB and only one of ethernet/sound (quad-gige in MPC8640D case, or sound in the case of MPC8610). I know this sounds a lot, but it's the reality, there is not enough funding to build enormous amounts of units and bring the prices down substantially, we have to start low and build up from there. In case you are wondering, yes, the boards will be designed/produced by bPlan and funded by my company (Codex).

About MorphOS support, well, I am not a user myself, but I would seriously consider funding it, if there was enough interest. If just a couple of users would be interested then the best thing I would do is send a working board to the MorphOS developers. the same goes for other OSes (incl AmigaOS). But I will definitely fund the Haiku porting, because I like it and I think it has a lot of future.

I would like to make a list of everyone that is really interested in such a system, so it would really help me make a decision sooner rather than later if you would send me a few personal details to markos@codex.gr with subject "PowerPC board":

* Name
* Country
* email (definitely, I'd have to reach you back!)
* Phone/Skype (optional, please include international prefix)
* Forum you saw this post (ok, Morphzone in this case)
* OS of preference
* board you would be most interested in (MPC8610/MPC8640D/P1022)
* preferred price (please have in mind the estimated price quotes I mentioned, it might be lower but that's not very probable)
* Other notes/comments

I'm not filthy rich, and I'm not doing this out of a hobby, if I'm going to invest this money, I want to know that I will make a profit out of it). I would appreciate it if you could post a link of this post on other forums (amigaworld?) because I do not really follow every forum.

Thanks a lot for your time and I hope this system becomes a reality.

Konstantinos Margaritis
Codex

[ Edited by feanor on 2009/9/23 12:33 ]

[ Edited by feanor on 2009/9/23 13:39 ]

A nice idea and a project I'd love to do myself but of course don't have the money.
I would certainly be interested if it ran Morphos. But I think to get a significant number of Morphos/AOS4 users on board you'll need something with a higher clock frequency than 1Ghz. I know numbers don't tell the full story but it's a psycological thing.
From a Morphos perspective a microATX board with the single core MPC8610 and plenty of expansion possibilities would seem to make the most sense - a Pegasos II replacement. Morphos is coming to Mac mini so the small compact less expandable side of things is already covered.

Well, if these numbers are true:

http://freevec.org/content/libfreevec_104_benchmarks_updated

a MPC8610 (similarly MPC8640D and the P1022) are usually 2-3x faster than a G4 1Ghz, just because of the faster memory bus. It's not just a simple benchmarks, a kernel compile takes 1/3 of the time on the MPC8610 than the G4. Plus, USB2, SATA, PCI-e slots would give you an advantage over any Mac Mini. I'm pretty sure the MPC8610 would leave the Mac Mini in the Dust. The lack of Altivec in P1022, that would be a serious disadvantage, but it might be outweighed by dual core, builtin crypto unit and much lower price -a new system might cost 300-350EUR. I'm not saying I wouldn't like higher clocked CPUs, but that would mean having the price go up to 700EUR or even more. Eg. check the prices here:

http://gr.mouser.com/Semiconductors/_/N-5gcb?Keyword=MC8640D&FS=True

and you'll see why it's so hard to produce a high performing PPC system.
It would flop at such a high price. As a sidenote, I have to say looking at this stuff all this time lately, made me completely understand Genesi's decision to move to ARM. I just think there is a niche market left neglected, that of PowerPC enthusiasts, and this is what I'm trying to work on, being one myself.

But I do appreciate the feedback, thanks a lot. And everything will be taken into consideration when the time comes to make the decision.

Konstantinos Margaritis
Codex



[ Edited by feanor on 2009/9/23 13:40 ]

Well I thought the MPC86xx chips would be faster than the old G4's but I didn't realise they could be that much faster.
The extra cost for a higher frequency chip was about what I expected. I suppose 700 euros really is too much to expect people to pay. I only mentioned the MPC8610 in my last post because although the MPC8640D is a nice chip neither Morphos or AOS4 support dual core so for those OS's it would be a bit of a waste. Of course you're looking at Haiku so dual core makes sense there.
The low cost of the P1022 is appealing but no AltiVec and a wait (possible delays?) to release isn't so good at all.
The Pegasos II replacement type machine I mentioned in my last post is what I would do but hey it's not my money and I'm completely biased

Good luck with your project, it would be nice to see a new PPC board (of any kind) hit the market.

> USB2 [...] would give you an advantage over any Mac Mini.

Every Mac mini G4 has built-in USB2.

Quote:

right, I didn't know that, I stand corrected. In any case this is about a new system not a case against the mac minis. I ask for feedback on something, which I think will be better than most currently available hardware.

[ Edited by feanor on 2009/9/23 14:15 ]

Very, very impressive initiative. Konstantinos, just clone that MPC8610 evaluation board in you have at the smallest possible cost, that's all.

Those dual cores look sexy, and we all know multiple cores are the trend these days. But there's a better trend: Classy programming! Give certain people a clever, single core, AltiVec equipped computer, and the only problem will be asking yourself what good use you could give to such a nice computer. Because making a computer just for the sake of it is very educating (the kind of task that governments should fund!), but not a business.

Attempt the smaller standard form factor, because it's likely that people will like tu put your computer inside places more interesting than a desktop case.

And... go for it! If bPlan is behind, that's a guarantee in itself. Are Gerald and Thomas excited yet?

I would like to help you with economy feanor for such project. To get a confirmation from MorphOS-Team about this would help a lot too.

A new Pegasos machine would be great!

Quote:

Unfortunately that's not possible, for both copyright issues but also in terms of cost. Evaluation boards have lots of stuff that isn't really useful in an end product (like JTAG for example). Also they are quite big in size.

Quote:

No, I'm definitely not doing it for fun -well not only anyway. It pisses me off that there is no interest in whatsoever to support and promote the ppc platform, and I think many feel the same way. This is a market in itself. Not big, but I believe it's self sustainable, if the choices are right. For this reason, I believe I should look past things like AltiVec support and focus on more market penetration. If the price is right, other users who might not otherwise choose it, might think it a nice choice. But I don't know if a 500EUR MPC8610 board would be a right choice, as much as I like Altivec -and you know I do! Also I'll have to look to the future, 86xx/e600 cores are going to be obsoleted sooner or later, while the QorIQ series are the focus of Freescale right now -I know I wasn't really fond of them a while ago, but I studied the specs and I found out that they're very worth CPUs.

Quote:

Totally agree on that.

Quote:

They like the idea, and I already started discussions with them about the board. I don't think I'll have a second chance in this however so I'll have to make sure it succeeds the first time. That's why I wouldn't like wasting their time with constant discussions and constant specs changes. I'll have to make a decision about the hardware specs, the project's viability and then things will start rolling.

Again, thanks for the feedback.

Konstantinos

[ Edited by feanor on 2009/9/23 16:45 ]

Quote:

I'm pretty confident that if there is enough interest for MorphOS, I will make an agreement with the MorphOS developers that will benefit both. I admit I haven't contacted them yet, but I wanted to have things on a more concrete basis before.

Konstantinos

Personally I would prefer a 8610 @ 1330MHz over a 8640D @ 1 GHz over a P1022@1 GHz.
I think bplan should know pretty much about the 8610 already.
In principle I guess Freescale will shift entirely to QorIQ in the middle future, but before that transition is done and e600 gets abandoned I would like to see a new e600 system. And IIRC the QorIQ have some fpu weakness. They are no desktop processors, the e600 was designed for the desktop.
The 8610 is the easiest chip to start with, the price is fairly okay. The 8640D is not hat much more expensive, should be faster per core (bigger L2 cache) at the same clock. In future MorphOS has to cope with multi core cpus anyway, maybe the 8640D would be a good starting point for that endeavour.
I like small boards with integrated gfx, with the 8610 this is possible. For those not satisfied with the inbuild gfx unit, there will be the option to insert a gfx card into a pci-e slot. An 8610 system should provide 2-3 slots. Only one slot is too few.
The redtail design by NEC may be a good staring point to derive a small and pretty system from. I was thinking of getting in contact with them, but haven't yet.
In the end I am only interested if MorphOS will run on such a board.

Quote:

They do. AFAIK this one would be the faster to deliver than all three versions.

Quote:

So do I, but would you pay ~500EUR for one? This is the question here. Perhaps it will be less in the case of 8610, that will have to be seen when the exact parts have been decided.

Quote:

I don't disagree, but I have seen some early benchmarks that place the P1022 as slightly faster than the G4@1Ghz in absolute terms. Considering the much faster memory bus, it's definitely going to be faster allaround than the G4, at least where Altivec is not involved. This is why I think it's not something to ignore.

Quote:

I don't know the internals of MorphOS so I can't even make an estimate of how long this might take. But I still think it would be something that the MorphOS team would think about very seriously.

Quote:

In particular, the MPC8610 can include 1 PCI slot and 1PCI-e x8 slot (actually it can provide another PCI-e 4x slot, but this will probably be used on a SATA controller).

Quote:

They're not interested in anyone that's not willing to buy a few tens of thousands of units.

Quote:

I appreciate all the feedback, but I would appreciated it even more if you would send those comments in a mail, as I requested. It would make it easier to prepare a data sheet of all this info and see what the stats are and how I can use it make the right choices.

Konstantinos

With regards to MorphOS, the MPC8610 appears to be the most attractive choice out of the three that were mentioned. I think the key advantage of any new hardware would be the option of using graphics cards that include more than 64MB of memory. Irregardless of the CPU performance and potential warranty issues, this would certainly justify the added cost compared to second hand Macs.

The Mac Mini is a great machine in many ways, but the graphics memory is a bit limited when you are dealing with high resolutions and many (virtual) screens. A number of Pegasos owners already use cards with 128MB of graphics memory for this reason.

[ Edited by ASiegel on 2009/9/23 16:21 ]

Quote:

Sorry, I'm confused, did you mean the 8610 or the 8640D?

Prizes are lower on this page :
http://www.freescale.com/webapp/sps/site/prod_summary.jsp?code=MPC8610&tab=Buy_Parametric_Tab

(it's for 1000)

- Dual core CPU is useless in AmigaOS/MorphOS. So no 8640D.
- You can do several variants : low cost with 8610/800 MHz, up to 8610/1333 MHz.

Quote:

Prices for 1000 are not the same as for 100. Sure. We could probably get even lower prices than that, but the link was used to refer to the escalation of the price of the 8640D

Quote:

That's the wrong reason, imho, but I will note it down. And it's not useless, it's just not supported now.

Quote:

Well, true, but I'd have to do a spread out of the several models during production. I mean if there's lots of demand for the 1333Mhz models, then I'd be left with lower-speced 800Mhz ones, and vice-versa. But yes, that's a good idea if done properly, thanks.

Konstantinos

Sounds very cool to me :)
You got an email!

MPC8610 sounds like a good idea. On the other hand I'm not sure if it will be much faster than my 1.5Ghz Mac Mini. And another additional and big problem is the number of years that it would take to make MorphOS available for this board.

For the MorphOS community I think it's better to support Apple G5 machines. Some cores reach around 2.7Ghz and you can fit very nice graphic cards.

Don't get me wrong, developing a new board sounds very nice but it would have to be twice faster than a 1.5Ghz Mac Mini for people thinking about it (specially since just the motherboard would cost them more than twice the cost of a full Mac Mini).

Apple G5 machines reach 2.7Ghz and have 900Mhz memory bus. By the time you released that board most of people will be able to buy 2nd hand G5 hardware for very low prices.

I understand the difference between used and new but if you develop a new board keep in mind that it should offer more than 2nd hand hardware.

What about PPC 970 cpu family? BBRV stated the prices some years ago and the price was reasonable.

Is it possible to get some Cell cpus?

I would prefer PPC970 cpus... unless price is around 350Euros and performance twice than my 1.5Ghz Mac Mini.

[ Edited by Crumb on 2009/9/23 22:37 ]

[ Edited by Crumb on 2009/9/23 22:39 ]

I've sent the mail.

> IIRC the QorIQ have some fpu weakness.

Regarding the FPU implementation of the P1 (and P2) "weakness" is an euphemism.

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=6196&forum=11&start=32

@feanor
Great idea! It is really right time now to switch to a newer PowerPC based mobo from a Peg2 and I will definitely pay 500eur for a MPC8640D based system. But! Always there's at least 1 "but"! I definitely need more than 1 PCI slot, SATA or IDE doesn't seems to be a problem anyway coz there's a huge amount of IDE hw here, but I have a PCI sound card which I like and don't want to change to internal one, also I will definitely need an USB and I also have a TV tuner.
I'm using linux now much more than MorphOS because of my work and a poor development environment there, but if MorphOS will run there it would be a great thing.

> What about PPC 970 cpu family?

The PPC970FX is available between 1.0 and 2.0 GHz, the PPC970MP between 1.2 and 2.0 GHz. The even higher clocked parts (used in the PowerMac G5, PPC970FX up to 2.7 GHz, PPC970MP up to 2.5 GHz) were discontinued by IBM.

> BBRV stated the prices some years ago and the price was reasonable.

Actual Prices can be looked up at various distributors.

> Is it possible to get some Cell cpus?

https://morph.zone/modules/newbb_plus/viewtopic.php?topic_id=6218&forum=11&start=20#61964

@Crumb,

I too would like to see the MOS Dev Team go on to working on a G5 PowerMac port of MorphOS after they are finished with the MacMini and maybe after also finishing ports to a few Mac laptop models.

The advantage of doing it for the PowerMac is that we could use some already supported graphics cards, sound cards, etc. in the PCI slots of the PowerMac. I would hope that this would shorten the required time to port MOS to the PowerMac.

This of course only makes sense if there isn't other better new hardware that the team should decide to port to instead. I am not hung up on buying used Macs, but their price to performance ratio as far as PPC hardware is concerned seems to be the best bang for the buck. I think it is the best option for the short term future of MorphOS, but would not be upset to be proved wrong if something better at the same price comes along.

Quote:

Actually, I tried to look up some prices for PPC970 and couldn't find a distributor, would you mind pointing to some URLs?

Thanks

Konstantinos

I've sent the mail.