3 more NatAmi's - Yippee!

[HenryCase]

@ alexh
the NatAmi is basically promising 100% backwards compatibility

Wrong. They have never promised this, the only project that has promised 100% compatibility (or as close to 100% as is possible) is CloneA.

Think about it like this, is a stock Commodore A1200 able to run 100% of Amiga software? Even with hacks like WHDLoad, an A1200 will probably struggle to run some old OCS software (stuff that relies on Kick 1.2 for instance). There has never been (AFAIK) an Amiga that ran 100% of all Amiga software.

On top of this, backwards compatibility has never been the prime focus of Natami (unlike the CloneA), but rather the focus has been on pushing the Amiga architecture forward. Here's an interview (from March 2011) with a Natami team member that gives a more realistic assessment:
http://www.reviewmylife.co.uk/blog/2011/03/23/natami-native-amiga-interview/

7. How do you go about designing the chipsets to be compatible with the original Amiga chipsets?

We can’t give a compatibility percentage yet but any game that already runs on a 68060-AGA-amiga is very likely to run also on Natami. With WHDload there is already a project improving compatibility of the old games and this project also helps the games on Natami.

Kinda think all you "doubters" are deliberately missing the point in the NatAmi...

You're missing the point about the statements that have been made in this thread. The people who have commented aren't against Natami per se, but rather have disagreed with distorted statements about the Natami. It's the statements that are the issue, not the hardware.

Just like them, I like the Natami, and am interested in seeing it released, but get pissed off when people make stupid statements about the power it'll have, as it makes the project seem more ridiculous than it actually is. Tone down the hype and all will be well.

[HenryCase]

Erm... afraid your wrong there... when I first heard of the NatAmi last year and visited the site 100% backwards compatibility was clearly stated on it and is the main reason why I support this project so much...

You should support the CloneA then. The Natami team have never promised 100% backwards compatibility, if you know they did then share the website link.

This project may "seem ridiculous" to you and it's obvious you don't like it despite your claim to the contrary

Really? Let's look at it this way, offers to join the Natami team:
HenryCase: 1
Franko:

As I said before, I like the Natami, have supported it/defended it for about 3/4 years, and am excited about its release. What annoys me is the fanboys who make outlandish claims not based in reality (plenty of examples if you need them), which makes it harder for people like me to convince skeptics to take the project seriously.

I'm repeating myself now, if you can't understand that I don't have a problem with the Natami by now then you're not listening.

[HenryCase]

So now we are keeping score? Can you get any more childish?

Nope, that's pretty much as childish as it gets for me, but if you have someone doubting whether you support the Natami you can't get a bigger indication that this is false than being offered to join the team making it.

Just as a side note, activity on Natami forums has been a little slow all of a sudden (no new posts this week), perhaps it would be good for Natami fans to show your appreciation in a more direct manner, get some activity going again:
http://www.natami.net/knowledge.php?b=0

[HenryCase]

@Franko

Erm... here where it says "The Natami is compatible to the OCS/ECS/AGA AMIGA chipsets"

http://www.natami.net/amiga-compatible.htm

 So unless the term "compatible" is being used here to mean "ok it might be wee bit compatible and not exactly quite compatible" then I take that to mean it will be 100% friggin compatible...

Ah, I see where you've misinterpreted things now. Simply put, compatible != 100% compatible. Think about it like this, Minimig v1.1 has an OCS core, it's OCS compatible, but it's not 100% OCS compatible. The reason being, it's not the original chipset, it a recreation, a clone, based on available documentation and code.

The same is true with Natami. It won't include original OCS/ECS/AGA Amiga chipsets, it'll include clones of those chipsets.

Now I'm not knocking the Minimig or the Natami, I think both are great and will continue getting better, but I understand that achieving that fabled '100% compatibility' is a hard task. If you don't believe me, take a look at UAE, which has been in active development for 15 years. Even now, with the latest versions of WinUAE, you'll find minor fixes for games/applications. Part of the reason is that the documentation for the Amiga is incomplete, and even if we had the original design documents, this isn't enough, as there are undocumented features/bugs in the hardware that you have to emulate/recreate for 100% compatibility.

The interview I shared with you earlier gives a more realistic idea of what compatibility to expect. What I wouldn't want you doing is getting people excited about 100% compatibility on the Natami, only for them to find it's not the case when it's released. If I've burst your bubble of excitement then I'm sorry, it's not my intention, I believe you can still be excited about the Natami even if it doesn't run all Amiga software out of the box.

Big deal "you offered to join the NatAmi team"

I didn't offer, I was offered. PM from Gunnar back in March 2008, asking me whether I'd be interested in joining the team, helping out with Coldfire coding (back when they were considering Coldfire). I declined as didn't think my skillset quite fit (plus I didn't think I'd be comfortable dealing with all of the Amiga community in a diplomatic way, as this thread shows!), but took it as a recognition of my support (I helped keep the peace in early Natami threads on AW.net). So yeah, bit different from your Brownies example, but good luck with getting in! ;-)

You've got a strange way of showing your support, lets see... hmmm... you back up people who pop into a thread rattling their gums about other processor speeds that have sod all to do with the Natami, then you complain about folk saying how a happy they are about the NatAmi and tell them not to "hype" it up by saying things that have been clearly claimed on the NatAmi site and elsewhere...

You see disagreement, I see progress. The criticism in this thread has been fairly mild compared to what it was in the past. At the very least the criticism was focused on performance benchmarks. What I saw was that people saw elaborating on what the benchmarks mean as an attack on the project itself, which wasn't the case. I'm posting as a mediating influence to stop people making claims they can't back up to defend the project. The project stands on its own merits, it does not need to be called into question with dubious claims.

@whiteb

Minimig has *NEVER* had, nor will it ever have AGA, it simply has no room to put it in to the FPGA Bitfile.

I know people have been talking about FPGAReplay, and labeled it "Minimig AGA", but that was more built as a device to run Arcade Roms, it just so happens that because of the FPGA size, they have got a preliminary AGA chipset for the minimig core ported to it.  But again, due to the size of the Bitfile, it will NEVER fit in to the FPGA used in the Minimig, not only the size, but the number of I/O lines in the smaller FPGA chip too.

There are two (equal) definitions of Minimig:

1. The physical board known as Minimig.
2. The FPGA core known as Minimig.

Both are equally correct. Therefore, it is inaccurate to say there is not a Minimig AGA, as there clearly is (using definition 2).

[HenryCase]

You can get it *CLOSE*, but if you want 100% compatibility, then use a REAL amiga.

...or a CloneA (if/when released).

Yeah thats right, Definition 2 (At least the AGA version), does not run on Definition 1.

Yes. My point is that 'Minimig AGA' is not an inaccurate name for what it is, it perfectly describes what it is. Consider this, when Dennis was writing the Verilog for the Minimig before the first Minimig PCBs were produced, he still had a Minimig when he ran the Verilog HDL in a simulator. The name belongs to the HDL core, it just so happens one of the early Minimig platforms shared its name.

[HenryCase]

Well, I have not read what methods Jens is using to reproduce the AGA chipset, you can use Technical manuals and write approximations in to VHDL, or you can use Reverse Engineering techniques, where you can use solvents to dissolve the ceramic of the chip and gain access to the core.  Either way, It is still using FPGA, but Jens is likely to have more accurate timings between the custom chips in his version..

Logic analysers are the main tool being used, from what I understand. Idea is to recreate the chipset(s) in a cycle-exact way, so that they behave in an identical way to the originals.

Jens has only so far announced CloneA to implement OCS (That I can see).

Not quite. Indivision ECS has HDL code from CloneA, and a completed AGA implementation has been hinted at, at least. See here:
http://eab.abime.net/showpost.php?p=712136&postcount=14

However, it is true that the OCS core is the only one to have been publicly demonstrated. For all those who haven't seen it already, torrent link for download here (used to be on Google Video, can't find it anymore):
http://breakpoint.untergrund.net/torrents/BP07_Seminar_JensSchoenfeld_Amiga_XViD.avi.torrent

I still think FPGAReplay is the one to watch.

Whilst I think interesting projects are happening all across the Amiga landscape, FPGAReplay is a great project, no doubt.

[HenryCase]

Back on the topic of Natamis, saw this video on YouTube the other day, think it would be a good target for Natami to aim for to show off its performance:
http://www.youtube.com/watch?v=02ZJNz3kt9o

If Quake 2 runs that well on a 68060 (admittedly with a Voodoo 3 card too) on a normal Amiga, I imagine you'd be looking at comparable (or possibly better, with the increased memory performance) results on a Natami. Worth a go at least, right? :-)

[HenryCase]

so why do folk always have to judge a machine by how well it can play one of these poxy ruddy games...

People use these types of games as yardstick of performance as they're fast paced action games, open-source and ported to many platforms that (in the case of the Quake series) have framerate capturing tools built in (maybe Doom does too, not sure).

If you know of a better game to measure 3D performance, please suggest it.

[HenryCase]

Who cares about 3D performance on an Amiga... :confused:

The idea is to test out the limits of the system.

It's certainly not in question that 2D performance will be fine (should exceed NeoGeo 2D performance, possibly even Saturn 2D performance, and I doubt any new 2D game made for the Amiga is going to reach the graphics standard of games on those systems). The more interesting question is what 3D performance will be like.

If that doesn't float your boat, fine, but it's interesting to some of us.

[HenryCase]

Is it possible to use a better benchmark?

What do you suggest we use instead?

Dry numerical benchmarks may be more accurate, but they don't capture the imagination in the same way. A demo coding competition may be an option, but that only really applies after the hardware has had a full launch. A launch title or two is an option, but they're unlikely to stretch the hardware without extensive effort being put into them.

In my mind that only really leaves one option, which is re-using existing code. Do you agree? So what existing code is likely to be a better way to show off the power of the hardware than Quake 2?

[HenryCase]

I get it by now that your a fan of these type of games

Actually I'm not.

You keep asking people what do you suggest but can only come up with the same answer time and time again

I keep suggesting the same answer because nobody is offering anything better, and it's the best one I can think of.

A web browser, not bad, but not terribly impressive either. Remember, we're looking for something to show of the power of the system, a web browser is a useful tool but I rarely go to a webpage and get blown away by the graphics.

So what would be more impressive, graphics wise, than Quake 2 (and yes I know I said that name again, let's see if you can respond reasonably)?

[HenryCase]

Let me guess... erm Quake 2 by any chance... :rolleyes:

No flies on you, eh.

Like it or not, the main draw of the NatAmi is SuperAGA (that's what sets it apart from the FPGAArcade). Reading through posts on natami.net, the impression I get is that the graphics side of the chipset is where most of the improvements are being made at the moment. To help sell the NatAmi it makes sense to showcase the area where it offers the biggest improvement over the competition, do we at least agree on that?

From what I remember of your previous posts, you didn't/don't use your Amigas much for gaming. Nothing wrong with that, we're all free to use our computers how we wish, but I do wonder whether part of your assertion that I have a 'teenage boy racer mentality' comes out of this.

My concern is with showing NatAmi off, to generate more interest. It doesn't have to be competing with cutting edge devices, but it does have to be interesting, and when I say interesting I mean in a 'I never expected a retro machine to be able to do that' kind of way.

Your suggestions were a web browser and a paint program. Amigas already had great paint programs, plus to make the most impact you'd have to find talented artists to produce work using it, and that art needs to be noticeably better than the great work already produced on Amigas. How do you intend to make that difference noticeable? More colours?

As for the web browser suggestion, let's say we're looking at a faster, native port of Netsurf, which is probably the best option. Take a look at this screenshot of the native RiscOS version, and tell me honestly, are you interested in buying the machine it runs on?:


Don't get me wrong, a web browser is definitely going to be a useful tool in getting people to buy, but it doesn't push the machine to the limit in a way that's truly going to grab people's attention. It's an essential, not something special.

With the estimated price a NatAmi is going to cost how many folk who are really willing to pay that price are going to do so just to play a 3D style game that they could do so better on a higher powered and cheaper PC or games console...

The idea is not that they'll buy a NatAmi to play an old game, but rather that it'll inspire them, knowing how much power they've got their hands on, to do great things with the machine. Think about when you first saw what the Amiga could do, the focus wasn't on mundane applications like spreadsheet and word processing software, it was on the power of the machine; the graphics, the sound, the fast multitasking, etc...

Believe it or not, I'm not tied to one game, and I'm trying to encourage some creative ideas, but they have to meet the criteria, and the only criteria is that the software must push what the machine can do in a noticeably impressive way. Think about if NatAmi got some magazine coverage, what would you want the screenshots to show?

[HenryCase]

Erm... slight problem... how we gonna write demos to take advantage of it's features until we own one... :confused:

Exactly! I already stated this option and dismissed it for the same reason. However, once the machine is properly released I have big hopes it'll get used in demo competitions.

[HenryCase]

Alright, we we find one of the best demo coders in the scene and offer him a beta board if he'll code for it.

The NatAmi is still work in progress, they'd have to wait until it was finished before releasing a hardware-banging demo, otherwise the smallest change in functionality could break it.

[HenryCase]

I find that unlikely. It couldn't compete with classic amigas since it'd be unfair. So the only category would be some "wild demo" or so. Also, more than often demo coders want to release their demos for platforms that are widely available.

Sure, you wouldn't compete them directly against other Amigas, and the first demos for it probably would end up in the "wild" category, but hopefully it could develop into its own category. I'd say many people still look at the years when Amiga was at its peak as the golden age for demo coding, this nostalgia plus the chance to expand on an existing skillset should (hopefully) prove compelling to some in the demoscene. It's not guaranteed, but it's not out of the question either.

At least one high profile & active amiga scener has already voiced his disinterest.

That's just one person though, it's hardly indicative of a trend.