NatAmi 68070 design draft

[amigadave]

bloodline wrote:

Crumb wrote:

Something as complex as a 68k CPU is never going to get even 20Hmz real speed on an FPGA... I'd be surprised if that to be honest...

Tobiflex m68k core is faster than a 20Mhz 68000 with a DE-1 board.

Yeah, and really how fast does it actually run?

Well with all that positive energy, it is sure to do well!

I am hoping that the NatAmi team does not give up when they come across a few tough obstacles and is able to prove the "Nay Sayer's" all wrong and come out with an improved Amiga 68k design that does indeed run faster than any existing Classic Amiga.

Why spend so much energy telling people that their plans are impossible when such a project would be a benefit to the Amiga community?  

And, for the person who wrote that no one in the Amiga community wants a super 68k and super AGA, WRONG!  That is exactly what I would like to have.  A 200-400mHz 68040/68060 with Super AGA or AAA that is fully backward compatible with AGA and ECS, OCS software.

You can have your PPC AmigaOne and OS4.0, I would rather spend my time working on and with OS3.x and what might be possible with the NatAmi.

As a commercial project, it may not succeed as there are not enough people that will purchase it, or enough developers that will write programs for it.

I'm too tired to rant any further.

[alexh]

bloodline wrote:

Crumb wrote:
Tobiflex m68k core is faster than a 20Mhz 68000 with a DE-1 board.

Yeah, and really how fast does it actually run?

What do you mean? I would imagine it runs at 20MHz. Although MHz is never a good comparison. You need something like the SPECint performance to get a good comparison. Several of the instructions in TG68 are many times faster than a regular MC68000 meaning a 20MHz TG68 is probably more than 3x faster than a 7MHz MC68000

[a-pex]

The natami developers are living 20 kilometers next to me ;-)

[Atheist]

bloodline, you funny guy.

You know, Amiga IS superior.

We MUST have the NatAmi60 because:

First; genuine RAM: disk. Then, genuine draggable screens.

Oh, and what of RAD: and it's surviving reboots?

And I can't install winXP with SP 2 in the ram disk of WinXP (which doesn't exist) and reboot winXP from ram. So clearly winXP is an INFERIOR OS.

xp doesn't have those, and can't operate with swapspace deactivated, and it DOESN'T tell you that that's why it won't work, it just doesn't (this is NOT a myth). Then, if you tell the computer to NOT accept input from the ethernet (internet connection), my computer GRINDS to a halt, well, REALLY slows down, anyway.

MP can kiss my ass. I want a ONE USER OS, and bill gates, deceptive rat fink, can NOT tell me otherwise.

Software architect nothing, he couldn't write a notepad for AOS2.0 replacement that WORKS.

My computer of 2.26 GHz P IV with 1 Gig ram stalls (~3 seconds) when opening a 2 to 4 meg ASCII file!!!!!!!!!

AND when I save the smallest trivial web pages with FireFox 2, it take 5 to 10 seconds (not kidding, SECONDS) before I can use FF again.

HE decided people wanted colour, then HE decided they wanted  GUI then HE decided people need MP, then HE decided you want registry and multiple logins, hey now, play games TOO......

WE HAD ALL THAT!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
(bar logins, registry, MP (but MP WAS optional))


The AOS1.3  more command, I couldn't press the space bar fast enough to outpace more reading off a FLOPPY DISK!!!!

xp is PURE GARBAGE.

[Atheist]

Okay, what do I know, I'm not that techie, but what about EPROM and EEPROM?

[Atheist]

And what about DLLs and notepad on xp being 64 K in size?

Our programs are mind bogglingly SMALL and yet VERY feature filled, and xp is 32 bit, and so is AOS!!!!!

We can multi-task without MP and self modifying code IS permitted.

It's XP that lacks FEATURES!!

[alexh]

Atheist wrote:
My computer of 2.26 GHz P IV with 1 Gig ram stalls (~3 seconds) when opening a 2 to 4 meg ASCII file!!!!!!!!!

AND when I save the smallest trivial web pages with FireFox 2, it take 5 to 10 seconds (not kidding, SECONDS) before I can use FF again.

Sounds like you have user error. Probably too much CRAP installed as background processes, probably some rubbish AV software monitoring everything you open, close, save etc.

50Mbyte ASCII file takes about 3 seconds to load here.
After saving webpages with Firefox I can use FF instantly.

[Piru]

@alexh

You can't reason with this guy.

[alexh]

Plaz wrote:
I imagined they were trying to do it all in one [Coldfire] CPU, and that can only handle so much.

And you want to have more than one?? Does the Coldfire have any hardware multiprocessing capability? Even if it does, it will be a nightmare to develop and write software for unless you abstract the entire system... and if you are going to do that... why use Coldfire?

[Crumb]

We MUST have the NatAmi60 because:

First; genuine RAM: disk. Then, genuine draggable screens.

 :roflmao:  :hammer:  :smack:  :roll:  :crazy:  :roflmao:

Oh, and what of RAD: and it's surviving reboots?

RAD is mostly useless for me. And it's useless for most of people with a stable OS that doesn't reboot each 10 minutes.

And I can't install winXP with SP 2 in the ram disk of WinXP (which doesn't exist) and reboot winXP from ram. So clearly winXP is an INFERIOR OS.

Erm... yeah! sure!

 :roflmao:  :hammer:  :smack:  :roll:  :crazy:  :roflmao:

xp doesn't have those, and can't operate with swapspace deactivated,

There's something wrong in your system because I can. Working without swapspace is not a myth, I have done it in the past (it's not a good idea since running out of ram is worst than having some slowdown when paging occurs)

My computer of 2.26 GHz P IV with 1 Gig ram stalls (~3 seconds) when opening a 2 to 4 meg ASCII file!!!!!!!!!

Your computer is badly configured by the user.

AND when I save the smallest trivial web pages with FireFox 2, it take 5 to 10 seconds (not kidding, SECONDS) before I can use FF again.

You have rubbish installed in your system, don't you? Norton antivirus perhaps?

xp is PURE GARBAGE.

Perhaps, but your reasoning for claiming that is totally flawed and insane. And it's pretty clear that you don't know how to use or configure your peecee.

Perhaps you should switch to AROS. No m68k apps but AmigaOS-like OS. Oh! I almost forgot you hate everything that doesn't have a label with "the name" ;-D

[Crumb]

bloodline wrote:

Crumb wrote:
Tobiflex m68k core is faster than a 20Mhz 68000 with a DE-1 board.

Yeah, and really how fast does it actually run?

Using Tobiflex "FAMPIGA" m68k core on a DE-1 board SysInfo reports a 43.10Mhz 68000. Since it's not a JIT, there are no caches involved and the timing should be quite accurate you can be sure it will be always much faster than a standard 68000.

Is there any brave FPGA coder who wants to include 68020 instructions to Tobiflex "FAMPIGA"? :-)

[jj]

@ Atheist

OMG!!!!!!

Really if you are going to post, try not to make youself look like such a complete idiot.  You obviously have no idea whatsoever about anything.  You obviously do not know how to operate or set-up any comwputer.  I wouldd dread to think the mess you would make of any OS install.

[wolfchild]

Whenever Natami is mentioned, it seems inevitable that the project team members will eventually be ridiculed.  I've been watching threads on other sites and I already can guess how it will end here as well...  No wonder they prefer to discuss only on their official site!

Ok, maybe certain design aspects are being under-estimated, but with all the 'I know better' style of comments is this the right way to encourage them?  Or do we have here a bunch of people who find delight in seeing others fail?  

Without knowing how they are implementing the system, who are we to judge about clock speeds and such?  Maybe they are doing a parallel design, and obtain 200MHz equivalent speed through parallelism?  Maybe they are basing work on TG68?  Who knows?

Their idea of using a custom 68k implementation is that they can tweak caches to snoop the chipset bus/registers or something like that.  Whatever that means, it seems like a good idea to me, and NO ONE should be telling them that it's not worth the trouble.  They will do their tests and they will decide whether to pursue that or not.

Whatever it may be they're doing, I believe that lots of people on this forum can provide great input to the developers about certain specs for the system, rather than criticising their decisions or claims.

Now that I got this off my chest, I'll just enjoy the whole Natami show as a spectator, and who knows, maybe I'll buy one as a toy someday.  Apart from that, this thread has become quite amusing lately thanks to fan #1...

[alexh]

wolfchild wrote:
Whenever Natami is mentioned, it seems inevitable that the project team members will eventually be ridiculed.

In order for them to be ridiculed, first they must write something ridiculous surely?

wolfchild wrote:
Without knowing how they are implementing the system, who are we to judge about clock speeds and such?

15 years of ASIC and FPGA design and a current job in a top chip company??

wolfchild wrote:
NO ONE should be telling them that it's not worth the trouble.

No, but at the same time someone should be telling them to be realistic. If they have their heart set on a soft 68k core then they should progress along the development. Start off simple and progress the design forward. Evolve their design. Not make any wild claims about prices, speeds, and structured ASIC's until they are well into their testing phase.

I wish their "team" all the best with their development. I hope that they can inspire and collaborate with the "do-ers" in this area, Tobias Gubener, Arnim Läuger, Gary Becker, Wolfgang Förster, MikeJ and the people over at PACEDev (and everyone else I missed) into taking part in debates.

[RedskullDC]

Hi Crumb,

Is there any brave FPGA coder who wants to include 68020 instructions to Tobiflex "FAMPIGA"? :-)

We toyed with this idea once before:

http://www.amiga.org/forums/showthread.php?t=32100

Probably not TOO difficult to make a 68020 "CODE COMPATIBLE" version, but with 16bit data bus and minus 68020 goodies like:

256byte caches.
three-stage pipeline.

Odd-word aligned data is probably the biggest enemy, though most high-level compilers are unlike to produce such code.
Normally there would be a compile option to avoid producing odd-word aligned data anyway.

Red

[LoadWB]

A lot of negativity here.  Some people miss that many technological advances and some commercial successes started out as an idea, a crazy notion, or a fanatical hobby.  While NatAmi may not be a way to compete in today's market, we should not and cannot discount that it may spawn other ideas or advanced methods which develop into better products.

A long time ago at a C64 swap party I mentioned that I wanted to hack together a real 6551 and support software and give the C64 the ability to communicate faster than 2400 baud.  I was dissuaded by all in the group because "the disk drive won't even run that fast."  A few years later CMD introduced the SwiftLink-232.

I should have done it, anyway.  And though it looks as if they do not really need it, I give that advice and encouragement to the NatAmi team.

And to the guy who mentioned self-modifying code, that is a good idea and all for old-timer programming (like 6502 hacking, demo stuff, etc.,) but in the modern world it is a massive security risk.  Apparently modern programmers and hardware architects have a pretty good grasp on that, now they just need to find a reliable way to prevent stack and heap corruption from pwning b0xen.