Oops. I stand corrected. The X1000 is only 6 years behind in terms of performance. By the time it's released, it'll be 7.....or more.....if it's ever released.

Hope is not a plan. I'd like to see an Amiga that can use Nvidia's GTX 590, but it isn't going to happen. And nothing about Amigas these days is "low cost". A-Eon wants something like $3000 USD for 6 yr old technology. And the SAM is ridiculously overpriced as well. As a hobby OS, the Amiga is doomed to underpowered, overpriced hardware that only the hardcore fans and dreamers can afford. The days of a cheap, powerful Amiga system have long since passed. This new SOC won't change that.

Last edited by ferrels on 27-Jun-2011 at 03:48 PM.

Quote:

This new SOC won't change that.

It would if they made it in high numbers at $20 a piece for the CPU. Of course now there are less Amiga people interested in new hardware, so this type of thing needs to be done at the start.

Even if they could have made 10,000 AmigaOne's at $100 a piece, would it still have been a bad idea if the hardware was buggy?

This would be a good future low end AmigaOne, but, IMHO, it needs to be significantly cheaper than 460 or don't bother at all.

_________________

@ferrels

Quote:

Oops. I stand corrected. The X1000 is only 6 years behind in terms of performance

Performance per Watt of PA6T is on the level of 2007/2008.

Quote:

The days of a cheap, powerful Amiga system have long since passed.

Even A1200 was outdated in its time (slower CPU, low resolution GFX), so this time passed really long ago.

Quote:

And the SAM is ridiculously overpriced as well.

Sure, comparison to x86 world is not plausible. However, in comparison to other "undead" platforms like Atari (FireBee) or Risc OS (ArcMini) SAM looks really good (and cheap).

Yes, price of X1000 is horrible, but there were computers in the Amiga history with much worser price/performance ratio (eg. A4000T from Escom). I will buy X1000 for any price - it is only money.

Quote:

Performance per Watt of PA6T is on the level of 2007/2008.

If I was in the business of building a laptop where battery life was a consideration, performance per watt would be important. But the average user isn't concerned about laptop specs applied to his Amiga desktop system. They're more concerned with raw power and benchmarks to back it up such as MIPS, specint, etc...And you'll never convince me that the SAM is cheap. I can buy an outdated PPC Mac or an x86 system for a fraction of what I'd pay for a SAM and it will run rings around the SAM, even in the price per watt ratios. Good luck waiting for your X1000 at any price. A-Eon has missed every deadline that they themselves have set. Be prepared for more missed deadlines and the eventual announcement that the X1000 will never be released. I'm sorry to sound so pessimistic, but I've been around too long and have seen too many Amiga NG fiascos. A-Eon has done very little to convince anyone that they'll successfully produce anything. Images of one working prototype running Debian and some possibly doctored shots of it running OS4 don't instill a lot of confidence in me or quite a few of my colleagues.

@ferrels

Quote:

I'm sorry to sound so pessimistic, but I've been around too long and have seen too many Amiga NG fiascos. A-Eon has done very little to convince anyone that they'll successfully produce anything. Images of one working prototype running Debian and some possibly doctored shots of it running OS4 don't instill a lot of confidence in me or quite a few of my colleagues.

You seem to have forgotten that A-Eon publicly demonstrated A1-X1000s with AmigaOS 4.1 running on them. There's even video evidence of that fact.

You are being incredibly pessimistic, and unnecessarily so.

Hans

Last edited by Hans on 28-Jun-2011 at 05:41 AM.

_________________
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
https://keasigmadelta.com/ - More of my work.

By using a SoC there is far less room for buggy hardware like with the AmigaOne.

This particular SoC isn't too bad at 1.2GHz with USB3, SATA, PCIe, WiFi - but it is clear it is meant for more embedded applications - see the 32-bit memory bus for example.

However it is clear that they'll bring the new hardware from this product into their higher-end products soon, so there'll be dual-cores with 64-bit memory buses and USB3, etc.

The only thing you'll need to add is a graphics chip (a low end embedded Radeon or Geforce would suffice) and RAM. No need for £500+ motherboards here.

@ferrels

There exist people that care how much electricity their desktop consumes.
For example if I have my computer on 24h/day & 365days per year, it is 50€ cheaper to have a SAM than AMD64x2(3800+). And SAM is the faster one of those two, except for things like mpeg2-mpeg4 conversion. And with AOS it is nicer to shut down the system and have it up & running very fast, with ubuntu it now takes minutes to boot the AMD box up.


and about the time it takes to produce a new PPC board...
For ACube it seems to take about two years.
(x1000 has some weird things like SB600 adapted to PPC and the xcore subsystem)
So for a new PPC board (simpler than those SAMs with embedded features) it shoudl not take two years to get the HW in production.

And also for the x1000 ... the work of getting PPC based AOS to run (at all) on a Power core based CPU took also some extra effort (not to mention ASMP + 32Gb memory address space + ....). One day, after AOS4 is ready for PA6T, supporting 465 (Power) core based product seems slightly simpler.

Last edited by KimmoK on 28-Jun-2011 at 06:00 PM.

_________________
- KimmoK
// For freedom, for honor, for AMIGA
//
// Thing that I should find more time for: CC64 - 64bit Community Computer?

Quote:

I think performance PA6T is similar to Pentium D (or probably faster in some cases) - Pentium D was released in 2005.

I'd be more inclined to put it at Core Duo / Core 2 Duo type level.

On integer based stuff (ironically, something SPECint is notoriously bad at measuring). it'll be more towards the Core Duo end.
For SIMD (AltiVec) type stuff it'll clearly be into Core 2 territory.
For memory it's another story altogether, that's at Core i7 type levels.

So it'll not rock at web browsing but it will be pretty good for watching movies or processing audio.

That said it's better than any current solution from any camp (except AROS).
I'll be very surprised if it doesn't beat the G4 and Sams at *everything*.
I expect it'll even give the G5s a good run for their money in some areas.

_________________
Whyzzat?

@minator

Quote:

I'd be more inclined to put it at Core Duo / Core 2 Duo type level.

I´m rather nore conservative in my expectations.

@Hattig

Quote:

The only thing you'll need to add is a graphics chip (a low end embedded Radeon or Geforce would suffice) and RAM. No need for £500+ motherboards here.

Please no. No low end graphics chips either on a desktop motherboard, or in a laptop/netbook. With a desktop motherboard, just give me a PCIe slot and let me choose what I want. If anyone is up to designing a laptop/netbook, use the MXM interface standard so that I can choose what graphics chipset I put in there. Even if you decided to bundle an MXM card in, we'd probably still end up with a better card than if you hard-wired one to the board.

Hans

_________________
http://hdrlab.org.nz/ - Amiga OS 4 projects, programming articles and more. Home of the RadeonHD driver for Amiga OS 4.x project.
https://keasigmadelta.com/ - More of my work.

Quote:

Poster: ferrels Date: 27-Jun-2011 19:58:07 Good luck waiting for your X1000 at any price. A-Eon has missed every deadline that they themselves have set. Be prepared for more missed deadlines and the eventual announcement that the X1000 will never be released. .

why don't you put your money where your mouth is and let's bet on it

_________________
____________________________
c64-2sids, A1000, A1200T-060@50(finally working!),A4000-CSMKIII
! My Master Miggies- Amiga 1000 & AmigaOne X1000 !
mancave-ramblings
X1000 I BELIEVE

Quote:

By using a SoC there is far less room for buggy hardware like with the AmigaOne.

IMHO Eyetech would have been betyter off with the original plan having the A1 as an A1200 accelerator or using the hardware as a basic for the A1 rather than scrapping it for something else made another company not up to scratch. But that boat has sailed.