@phoenixkonsole
I like fpga and this is not meant as bashing. My Arguments wehre About how to increase userbase. Tactic: Infiltrate, Inform/brainwash-> get a new amigan

Last edited by phoenixkonsole on 25-Jul-2014 at 06:33 AM.

_________________
AROS Broadway - AEROS - Aminux - AmiCloud - indieGO! Appstore - AmiWallet - VAN lossless video codec - AMC Amiga media Center -KrypUnite - LibertyNet - MinX - amigaNX

@Nameless

Quote:

How cheap? It'd need to be around Raspberry Pi levels of cost.

There are now 100$ Bay-trail SoC tablets.

_________________

@WolfToTheMoon

$100 would probably still be too pricey though.

There is also the issue of attracting developers. If any Amiga-ish device has an X86 (or probably ARM), the question is, would developers even bother developing anything Amiga-related, when Win or Android would make more sense?

Or how many folks would buy an Amiga-ish device, if all they were buying is an X86/ARM with an emulator? I expect many would just say, where can I get that JIT emulator and I'll run it on an existing device.

So I do think unique Amiga hardware is important, be it FPGA or ASIC.

I seriouly would like to hear the view of the "parties" involved in AMIGA.
Like Hyperion, Trevor, Cloanto etc....


My personal limited view is that the people involved in AMIGA
and making AMIGA hardware all prefer a business model
where they e.g. sell 1000 boards with $300 revenue per board.
Over a model to make only $6 per board and to sell 50,000.

The first option is a sensible business move - but it does not help AMIGA.
Only the second option has a chance to revive AMIGA.

Would any of the parties in AMIGA be at least interest in giving the lower profit option to revive AMIGA a chance?

_________________
APOLLO the new 68K : www.apollo-core.com

@BigGun

I do not think you will get any public answer about business strategy. If you want answer ask them directly by email

@Nameless

+1

It is partly a emotional thing. A platform is only taken serious as a platform if there is specific hardware. When there are options to run the same software on other platforms and get more customers that is helpful but a hardware is needed. And FPGA are (at least for now) the most realistic option.

@BigGun
AmiQube Comes with licensed roms.

_________________
AROS Broadway - AEROS - Aminux - AmiCloud - indieGO! Appstore - AmiWallet - VAN lossless video codec - AMC Amiga media Center -KrypUnite - LibertyNet - MinX - amigaNX

@Nameless

There is a 60$ 7" Windows 8.1 tablet coming from Emdoor.

_________________

@Nameless

Quote:

There is also the issue of attracting developers. If any Amiga-ish device has an X86 (or probably ARM), the question is, would developers even bother developing anything Amiga-related, when Win or Android would make more sense?

Well, you need to make it interesting for developers. As we're witnessing, current crop of Amiga-inspired OSes do not offer anything to the world of outside developers/users.

_________________

@WolfToTheMoon and @Pascal

I think a FPGA based hardware is different and difference makes it interesting (more than emulation on ARM or X86 can). But again it sounds like there is only one possible route but I do not think that. The user base must be extended, there are certainly a lot of people who are more attracted by a FPGA based solution than by ARM + emulation (or X86/X64 + emulation). On the other hand there are also many people who would prefer a system (f.e. ARM) + emulation. Both are different groups and that split is also the case for developers. As long as the software runs on both (or is at least possible to write it in a way that it runs on both) I do not see a problem there. Why doing only one if you can do both

I also think that a standardized development and distribution platform is very important for any commercial development. That was the idea and hope I had when I started to work on my 68k distribution. I think it is less important for users but very important for developers.

Last edited by OlafS25 on 25-Jul-2014 at 10:02 AM.
Last edited by OlafS25 on 25-Jul-2014 at 09:45 AM.

@OlafS25
Yep I agree

_________________
AROS Broadway - AEROS - Aminux - AmiCloud - indieGO! Appstore - AmiWallet - VAN lossless video codec - AMC Amiga media Center -KrypUnite - LibertyNet - MinX - amigaNX

If we're discussing about Amiga FUTURE, than I see no point in 68K discussion and whether to emulate in software or with FPGAs.

Amiga future must include things that are relevant to potential customers today and tomorrow, and perfect emulation of OCS/ECS/AGA or SuperAGA is not. That is, if you want to expand the userbase...

More attention must be given to software/OS, and less to making obsolete overpriced hardware.

Last edited by WolfToTheMoon on 25-Jul-2014 at 10:22 AM.
Last edited by WolfToTheMoon on 25-Jul-2014 at 10:20 AM.

_________________

@WolfToTheMoon

sorry to say that, what has NG managed to get over the last 10+ years?

And even if there would be a super new amigaos (or MorphOS/AROS) that has 64bit, SMP and MP do you seriously think it would magically change the situation? Then it would offer the same as Windows/Linux/Mac but minus modern development environments and minus software base. Do you think users would change then and developers support it? I think it would not immediately change anything.

ARIX might come next to it so we will see then but I predict no flood of new software, perhaps some amigans will use it at least additionally but outside the community it would be a hard sell. Most of the codebase IS 68k and you underestimate the meaning of feelings (you could say irrationality) in this case. I personal became interested in amiga again when I found the natami site. A amiga successor with changable hardware that was really interesting. If I had only found AmigaOS or MorphOS or a site about classic gaming I would have said "past" and left. And I predict that I am not only one who thinks (and will think) this way. So only because you think different it does not mean everyone thinks like you.

And Amiga is about hardware (not OS). That people just talk about OS is new and different to the past. People outside used A500 (or A1200) so they will think about custom chipsets and 68k processor, not about expensive PPC hardware or used Macs or PCs with Amiga installed. I will bet with you that you will get 100x more publicity with new standalone FPGA than with updated NG OS.

And where do you get your "obsolete overpriced hardware" from? FPGA based cards are pretty modern, in fact they are used all over the industry and because of mass production they are cheap.

Last edited by OlafS25 on 25-Jul-2014 at 10:37 AM.
Last edited by OlafS25 on 25-Jul-2014 at 10:34 AM.

@OlafS25

Quote:

FPGA based cards are pretty modern

68060 300 MHz class hardware for 150 USD would have the same price/performance ratio as X1000 for 3000 USD.

@pavlor

Really?

Lets wait for the benchmarks first

Last edited by OlafS25 on 25-Jul-2014 at 10:47 AM.

@OlafS25

To me, 68K is past perfect tense and retro. I was under the impression this was a thread discussing Amiga future, not past.
50 000 users is doable with an NG-Amiga.
A C64DTV style or FPGA Amiga might sell nicely if cheap enough, but it wil not result in a renewed life of the 68K platform, just as the C64DTV did very little for the C64 codebase. People bought those mostly to play games - and I very much question the ability to sell something like that today when there are a number of extremely cheap ARM/x64 systems that can run emulation without breaking a sweat.

Currently, my most likeable idea would be to make something like Open Pandora, with x64, and run ARIX.

_________________

@pavlor

Quote:

68060 300 MHz class hardware for 150 USD would have the same price/performance ratio as X1000 for 3000 USD.

If you're talking $ per Mhz it's $.50(68060) vs $1.66(X1000).

If your talking real life performance. I guess we'll have to wait and see.

_________________
Support AROS sponsor a developer.

"AROS is prolly illegal ~ Evert Carton" intentionally quoted out of context for dramatic effect

@OlafS25

Quote:

Lets wait for the benchmarks first

68060 scores 255.000 in OGR-NG...
that is 1.530.000 at 300 MHz

PA6T (1 core) scores 23.000.000

OGR-NG is suited for multi-core operation (you can double that result with SMP capable OS like Linux).


Sure, dnetc is not ideal benchmark, but you can see performance gap between NG hardware and best classic CPU.

@OlafS25

Quote:

@pavlor Really?

No not really.

The problem with this discusison is that performance is a "soft-comparison" and that performance can be measured in so many different ways that you can "proove" everything if you wanted.

Example:
1)
A Multiplication on an 68000 took 70 cycles.
A multiplication on the 68060 takes 2 cycles.
= clock by clock the 68060 is 35 times faster

2)
A NOP instruction on the 68000 took 4 cycles.
The NOP on the 68060 took as far as I recall 9 cycles.
= clockm be clock the 68000 is twice as fast as the 68060.

Both above numbers are valid and true.
But both give back a very different "view" on the reality.




If you want to compare a system you really need to specify your "benchmark".
You can for example say the typical Sysinfo Memspeed values is a reasonable benchmakr for you.

Then you would see that old 68K system started with 2 MB and scaled up to 50 MB with a fast 68060.

Now a new 68K core could reach 400 MB /sec with the same benchmark no problem.
Which is about twice as much as GigagHerz AMIGA ONe G4 PowerPC does reach in the same benchmark.
So looking at this single benchmark your $150 FPGA AMIGA has twice the performance of the POWERPC NG Amigas.

_________________
APOLLO the new 68K : www.apollo-core.com

@WolfToTheMoon

CPU shouldn't even be part of the discussion anymore.

http://llvm.org/pubs/2006-06-18-WIOSCA-LLVAOS.pdf

_________________
Support AROS sponsor a developer.

"AROS is prolly illegal ~ Evert Carton" intentionally quoted out of context for dramatic effect