SILVER9 core released for Vampire 600 V2

[ShK]

Why, they can do them in parallelle. In my view, they need more hardware people, it will just take too much time if Majsta is to design all the boards, not to mention, it will drain on him. Another option of course, is that they license out the core (which I think was Gunnar's original idea) to people who want to use it on their FPGA projects.

With a stand-alone board you have to have all in features in operation at once. You can't get anything visible before everything is ready.

Now with acc cards you can add new features one by one and everyone can use them immediately.

In the end when you have all the necessary features tested and working, you can do also stand-alone board much faster.

[spudje]

A standalone board with USB that supports keyrah would be an excellent replacement for aging/broken motherboards. Just install in the original case and use the stock keyboard.

Just for my understanding, what would a standalone Vampire offer more than a MIST/Replay Arcade FPGA??

[freeaks]

Just for my understanding, what would a standalone Vampire offer more than a MIST/Replay Arcade FPGA??

the cpu used on vampire, is a next generation cpu in 68k family.
it offer full compatibility with all the previous cpu.
but, also it offer completely new instructions. multimedia instructions (similar to intel 'mmx') but also a simd (similar to ppc 'altivec') and many other new features.

not to mention raw power. apollo cpu performs a lot better than the cpu found in other standalone products.
sysinfo is a bad benchmark tool,
apollo team made 'minibench' which is more reliable.

vampire score 150+ mips

as you can see here:
http://www.apollo-accelerators.com/files/Apollo_datasheet.pdf

also, all the new intructions mentionned above, have to power to speedup even more data processing, for games and multimedia for example.

so, a standalone vampire would bring
- all backward compatibility
- a lot more power
- saga (think updated, faster aga)
- modern mainboard
with possibility to use off the shelves usb mouse, keyboard, ethernet, hdd ..etc.. without needing all the adaptors. recent chips, memory, capacitors, alimentation and battery.

[fishy_fiz]

Personally Id take any benchmark tool made by developers of hardware with a grain of salt. Throughout the ages theyve always favored the vendors hardware. Sysinfo may be unreliable, but its unreliable for all hardware, so theyre on an even keel.

Not that Im trying to dismiss the work done here. It's cool stuff that Im also interested in, but there's been a few things that seem to mirror the usual practice of hand picking/hand tuning performance benchmarks (like the weird suggestion that fpu isnt important, and to prove a point theyve shown interger performance of other hardware vs interger performance of apollo,.... how does this prove anything in regards to fpu? .

p.s. in regards to fpu, Im basing this on another persons post here at a.org,...... it's possible that more useful information has actually been posted elsewhere.

[QuikSanz]

A standalone board with USB that supports keyrah would be an excellent replacement for aging/broken motherboards. Just install in the original case and use the stock keyboard.

And toss all my Zorro cards? Nope.

[freeaks]

@fishy_fiz
everyone have been saying for years that sysinfo is not producing precise results. in addition to that, apollo have new instructions that sysinfo don't know about. sysinfo is just Not good enough anymore as a tool to get a good sense of a cpu's power.

nevertheless, apollo-vampire results on sysinfo are known.
it can be seen on the same link i gave earlier.

about fpu, it's coming.. a bit more of patience

[IanP]

I think it's important for the Apollo Core to become established as the natural successor for the Amiga 68K machines. The best way to achieve that acceptance is by supplying accelerators for the existing classic Amiga lines before trying to create stand alone products. Persuading an Amiga fan to add an accelerator to their beloved machine is easier than persuading them to purchase a new computer I would imagine.

[slaapliedje]

I've been waiting patiently for one for the A4000... will definitely order one when they become available.

[Kremlar]

I'd love a standalone to go along with my FPGA Arcade!

[AJCopland]

I think it's important for the Apollo Core to become established as the natural successor for the Amiga 68K machines. The best way to achieve that acceptance is by supplying accelerators for the existing classic Amiga lines before trying to create stand alone products. Persuading an Amiga fan to add an accelerator to their beloved machine is easier than persuading them to purchase a new computer I would imagine.

Then it'd be good for those who are able to offer to help them out designing boards etc.

It looks like a simple adapter for the V500 can cover the A1000/A1500/A2000, but that leaves A3000(/UX) and A4000D/T of the big boxes. Then there's the A1200 and CD32.

Part of the work is board design, layout, and routing which people could help with.

Go and volunteer over at http://www.apollo-core.com if you're interested and see what they have to say.

EDIT: I am not involved in the project, just suggesting that if people want to see a Vampire board for a particular machine, then helping to develop it might be the fastest route.

[Crom00]

Well said AJ, at this point it's up to the users. I don't think classic will supported by anyone out there aside from good folks like Jens and the Apollo Team and a handful of hardcore devs. So why not contribute. The SA version will be quite nice. I am a vampire team member. What I like most about the Vampire project is the whole Amiga /Atari Tramiel concept of power without the price.

[polyp2000]

Sysinfo may be unreliable, but its unreliable for all hardware, so theyre on an even keel.

I actually disagree here - its easy to game a known bench-marking tool when you know and have the ability to tweak to improve the results - (not that this is necessarily a bad thing). With regular 68k the statement holds, but not when you add an FPGA cpu into the mix.

Id be more interested in looking at how things run. Speed of rendering Lightwave scenes, emulation. Things that a regular Amiga would struggle with.

Real world usage is what counts - bench-marking really just gives you a general idea.

[psxphill]

JWhat I like most about the Vampire project is the whole Amiga /Atari Tramiel concept of power without the price.

Jack would never have used an FPGA, too expensive and upgradable.

[UberFreak]

A standalone board with USB that supports keyrah would be an excellent replacement for aging/broken motherboards. Just install in the original case and use the stock keyboard.

But then its not an Amiga anymore...

[OlafS3]

But then its not an Amiga anymore...

technical new standalone versions will be very different from old amigas

not amiga anymore?
Depends on definition

[Heiroglyph]

Oh God, here we go down this rabbit hole again.

The thread is about the new core, let's try to talk about that.

To me the price to performance ratio of the Vampire with the Apollo core is the best thing to happen to this community in decades.

I'm next on the list to get a v500 and I haven't been more excited about new hardware in years.

When the 4000 version or a stand alone is available, I'll be buying multiples and whether it's still an Amiga is the last thing on my mind.