Apollo Team has new Cyclone 5 FPGA accellerator cards

[jbenam]

Originally Posted by edd_jedi

I actually don't think it's got anything to do with 'wealth', spending a few hundred quid on computer equipment is within the means of 90% of people these days, it's more down to hype and/or stupidity in my opinion. You don't see old P133 processors selling for hundreds of pounds.

That's the thing - even if something is overhyped as hell you just don't waste 700 quids on it if you can't spare it.

The problem is that Amiga accelerators were a niche of a niche. While you didn't have as many Wintel machines around as you have now, they weren't exactly rare.

It's a matter of fact that the Amiga population around 95-96 was at best a mere 10% of the Wintel market, and of that 10% I'm not sure if even the 3-4% of it bought an accelerator for their Amiga.

That said, they aren't exactly rare either as of the thousands of classic Amiga users nowadays not many will spend 700 quid on an accelerator - so the few that are around are enough to cover the big spenders

Be assured that if a 060 costed just 150£ pretty much everyone would have one in their A1200, and the entire supply would just dry up in a few weeks

[edd_jedi]

Yeah I'd love one if they were more affordable, so here's hoping It's not that I can't afford one, I just can't mentally justify spending £700 on a 60mhz processor when I have a 2.8ghz i7 as my main computer that is now probably worth less :

[wawa]

Originally Posted by matthey

AROS needs to become more efficient.

bet, what i think about it. still with those few inexperienced testers like me and very few developers it "aint gonna happen any soon".

and and StormMesa can use software 3D in 128MB with no problem

storm mesa is mesa 3.0 right? aros mesa is mesa 7.x. current mesa is around 10.x. rhe demands grow with the capabilities of mainstream systems. i have talked about the memory footprint issue to deadwood, but
a) internal changes to mesa would be too difficult and time consuming, since there is no other platform that needs such measure.
b) the solution would be only temporary as it would affect an old mesa version and wouldnt be pushed upstream, so the problem would reappear with every new mesa port.
c) aros mesa is in comparison to storm mesa many times slower (in software). so it iisntz probable that it would ever become practicable solution especially on real amiga hardware.

The original target was 64MB of fast memory but I pushed for 128MB. I think it's the right amount for affordable accelerators. High end accelerators and standalone boards should get more memory.

i think 128mb if fairly enough when you take the genuine software base into account. but having better cpu will in turn increase demands on memory since new software will become practicable. for instance web browsing. currently the pure executable of owb takes 20mb, it takes ages to start it and it is rather slow in browsing, but imagine it would become popular with the fpga cpu users, who would actually start browsing todays css heavy internet with their amigas.

[Wiffleplop]

So we have a timeframe on the release of this board? I'm considering an accelerator from the ACA line, but if this thing is 3-6 months away, I'd be wasting my money on a short-term investment.

[trixster]

Same here, a timeframe would be good. I've been toying with upgrading from my Blizzard 1220 but the price of something worthwhile which is substantially faster, and the lure of 'announcements' like this, are stopping me taking the plunge.

[Mrs Beanbag]

Originally Posted by edd_jedi

I actually don't think it's got anything to do with 'wealth', spending a few hundred quid on computer equipment is within the means of 90% of people these days, it's more down to hype and/or stupidity in my opinion. You don't see old P133 processors selling for hundreds of pounds.

i don't know, people spend thousands on a new car that they only use for maybe an hour or so a day, yet i still baulk at spending £500 on something for my Amiga that i'll get far more use out of...

[gunni]

this is pretty epic news, I'd certainly snap one of these up

[Michael Sykes]

Originally Posted by Gunnar

and compared to an 68060 it reaches around 300Mhz 68060 performance.

That sounds like WinUAE power on real hardware.
WOW!

[Kehil]

Originally Posted by Michael Sykes

That sounds like WinUAE power on real hardware.
WOW!

Ehm..
This statement applies just to FPU part of planned FPGA core, if I understood well.

[ajack]

What I want right now is detailed specs, price and availability...

[cugar124]

I am with Ajack is this another promised setup that never sees the light of day seems like the only guy to actually make anything is Jens. Its shocking how many projects have failed. So the challenge is stop teasing and get it built Jens will never make anything above 030 because he knows no one else will ever make one to compete. Only completed project was the 600 vampire otherwise its just excuses.

[Gunnar]

Originally Posted by ajack

What I want right now is detailed specs, price and availability...

As I already said before:

The status is:

1) The Vampire600 works.
2) We use our new CPU core with it.
If you want to see the core running: http://www.apollo-core.com/bringup/onlyamiga.avi
3) We have designed a new card
The specs of the card were posted here before.


As I clearly said this whole thread is NOT an official announcement.

We will only officially announce the card as soon as our testing if finished
and as soon as we have the sales organized.

Until then this is just a sneak peak develoment discussion.

[Wiffleplop]

Understood - but you can't blame us for getting excited, and the ones that have been around longer than me for being a bit cynical. Six months? A year? Ball-park figures that we won't use in court against you is all we're looking for

BTW - loved the demo with the kids mouthing the words to the song

[Gunnar]

Card specs are:

* very fast 68K CPU
* 128 MB DDR3 Fast-memory
* SD-card usable/bootable as IDE-device
* Network interface
* RTG Graphics Card (chunky/Hicolor/truecolor) with HDMI out



Again: I can NOT promise any delivery dates.
We are too early in the testing now.

Our internal team plan is to have the card tested by the end of August.
As of today, I can give no guarantees that this plan works out.
We might be able to give a proper estimate in 4 weeks.


My advice : If you want to buy an upgrade now then just do it.

[Wiffleplop]

Thanks Gunnar - will have to see which comes first. My finances recovering from my recent splurge on retro-computing, or the release of the board! I kind of hope it's the latter really

[Vot]

Originally Posted by Wiffleplop

Thanks Gunnar - will have to see which comes first. My finances recovering from my recent splurge on retro-computing, or the release of the board! I kind of hope it's the latter really

Really? So you hope it gets released before you can afford to buy one? ...

[Gunnar]

Two questions:

1) We could sell the card as pure CPU upgrade earlier.
But then all what the users will get is very fast 68K CPu plus 128 fast memory.

Would this be an option for the people?


2) Would people be interested in something like blog which we create.
We could add an update every other day or post some pictures.
Are people interested in this?

[robinsonb5]

Originally Posted by Gunnar

2) Would people be interested in something like blog which we create.
We could add an update every other day or post some pictures.
Are people interested in this?

Most definitely - in fact this is by far the best way to manage peoples' expectations. No promises, no timeframes, just a window onto the project's progress as it happens.

[JimDrew]

Just a functioning fast 68K with at least 128MB of RAM would be fine for the initial release, with the hardware itself tested and verified that the video output, network port, and any other planned feature works correctly. Changing cores is just a simple file, changing/fixing hardware would be a nightmare.

[Samurai_Crow]

Originally Posted by Gunnar

1) We could sell the card as pure CPU upgrade earlier.
But then all what the users will get is very fast 68K CPu plus 128 fast memory.

Would this be an option for the people?

This sounds tempting! I'd make the chunky graphics mode last priority at this point anyway since if I'm coding on the Amiga I'd want to use AGA or ECS anyway. I still don't have money for upgrading my A1200 at this point though.

I've got a few questions and a comment:

Will the HDMI output in the final version support AGA chipset emulation as was originally planned for the SuperAGA chipset? Also, will there be sound passthrough in the HDMI header? (If not, there's no advantage over DVI digital other than having a smaller port is there?)

My current Blizzard1230mk4 has 16 megs of Fast RAM, SCSI2 adapter, FPU and full MMU installed. I might be tempted to sell it off and replace it with the FPGA board if ThoR could make a version of MuForce, MuGuardianAngel, Mungwall emulation from debug mode on AROS (Does that even use an MMU?), etc. that would run on it. It need not have a 68030 or even a 68060 compatible MMU since those are the only MMU utilities that would need to work with it. Those utilities are great for development and the main reason ACA accelerators don't sufficiently satisfy me or convince me to switch (not to mention that ACA doesn't go higher than an 'EC030 anyway).